National Clinical Guideline Centre

1

Draft for consulation, December 2014

Type 1 diabetes in adults Type 1 diabetes: diagnosis and management of type 1 diabetes in adults

Clinical guideline <...> Appendices A – F December 2014

Draft for consultation

Commissioned by the National Institute for Health and Care Excellence

Type 1 diabetes in adults

Contents

Disclaimer Healthcare professionals are expected to take NICE clinical guidelines fully into account when exercising their clinical judgement. However, the guidance does not override the responsibility of healthcare professionals to make decisions appropriate to the circumstances of each patient, in consultation with the patient and/or their guardian or carer.

Copyright National Clinical Guideline Centre, 2014

Funding National Institute for Health and Care Excellence

National Clinical Guideline Centre, 2014 Type 1 diabetes in adults

Contents Contents

Appendices A-F ...... 5 Appendix A: Scope ...... 5 Appendix B: Declarations of interest ...... 23 Appendix C: Review protocols ...... 41 Appendix D: Clinical article selection ...... 83 Appendix E: Economic article selection ...... 106 Appendix F: Literature search strategies ...... 107 References ...... 147

National Clinical Guideline Centre, 2014 4 Type 1 diabetes in adults

Scope Appendices A-F Appendix A: Scope

National Clinical Guideline Centre, 2014 5 Type 1 diabetes in adults

Scope

National Clinical Guideline Centre, 2014 6 Type 1 diabetes in adults

Scope

National Clinical Guideline Centre, 2014 7 Type 1 diabetes in adults

Scope

National Clinical Guideline Centre, 2014 8 Type 1 diabetes in adults

Scope

National Clinical Guideline Centre, 2014 9 Type 1 diabetes in adults

Scope

National Clinical Guideline Centre, 2014 10 Type 1 diabetes in adults

Scope

National Clinical Guideline Centre, 2014 11 Type 1 diabetes in adults

Scope

National Clinical Guideline Centre, 2014 12 Type 1 diabetes in adults

Scope

National Clinical Guideline Centre, 2014 13 Type 1 diabetes in adults

Scope

National Clinical Guideline Centre, 2014 14 Type 1 diabetes in adults

Scope

National Clinical Guideline Centre, 2014 15 Type 1 diabetes in adults

Scope

National Clinical Guideline Centre, 2014 16 Type 1 diabetes in adults

Scope

National Clinical Guideline Centre, 2014 17 Type 1 diabetes in adults

Scope

National Clinical Guideline Centre, 2014 18 Type 1 diabetes in adults

Scope

National Clinical Guideline Centre, 2014 19 Type 1 diabetes in adults

Scope

National Clinical Guideline Centre, 2014 20 Type 1 diabetes in adults

Scope

National Clinical Guideline Centre, 2014 21 Type 1 diabetes in adults

Scope

National Clinical Guideline Centre, 2014 22 Type 1 diabetes in adults

Declarations of interest

Appendix B: Declarations of interest

B.1 Stephanie Amiel (chair) GDG meeting Declaration of Interests Action taken On appointment Personal pecuniary: I have served on advisory boards Stephanie Amiel's declaration 13 January 2012 for Medtronic, Cell Novo, Novo Nordisk, MSD, Eli of interest in relation to Lilly. DAFNE was discussed with the I have received honoraria for lecturing/teaching Guideline Lead and it was from (or have spoken at meetings sponsored by) deemed to be reasonable for Abbott, Medtronic, Lifescan, Roche, Sanofi-Aventis. Stephanie to take part in the discussions. Several of the healthcare professionals on Non-personal pecuniary: I am co-head of a Clinical the GDG are trained Academic Group and an academic division in my educators. Wherever the institutions (King’s College London and King’s College interest was of relevance, the Hospital). They have received funding for research or GDG were reminded of it. service development from Novo Nordisk.

Personal non-pecuniary interest: I am a member of the scientific advisory boards of the Juvenile Diabetes Research Foundation and Diabetes UK I am on the editorial board for the Diabetes and Wellness Foundation. I am academic representative to the Executive Committee of the Association of British Clinical Diabetologists. Until last year, I was the Chairman of the National Dose adjustment for normal eating (DAFNE) Executive. I have published both as named author or as part of a consensus group, reviews, papers and guidelines concerning diabetes management and particularly hypoglycaemia in diabetes, including involvement in local/national/international guidelines on insulin pump therapy (including for NTAC); bariatric surgery in diabetes; use of glucose monitoring. GDG 1 Personal pecuniary: I append a list of advisory None 23 October 2012 boards attended and single (industry) sponsor educational meetings where I have been a speaker. I have resigned from all this work since my appointment to the NICE GDG chair position and have refused all subsequent invitations to such activities.

Personal family interest: My husband is Chairman of Diabetes UK and Chairman of King's College Hospital NHS Foundation Trust. I am not aware of but do not keep information on his other professional activities.

Non-personal pecuniary interest: I am co-lead of a Clinical Academic Group in Diabetes, Endocrinology and metabolism, Nutrition, Obesity, Vision and Related Surgeries (DENOVaRS) at King's Health

National Clinical Guideline Centre, 2014 23 Type 1 diabetes in adults

Declarations of interest

GDG meeting Declaration of Interests Action taken Partners (the Academic Health Sciences Centre comprising King's College Hospital and Guy's and St Thomas' Hospitals NHS Foundations Trusts, the South London and the Maudsley Mental Health Trust and King's College London). DENOVaRS is currently undertaking a project 'Changing Diabetes at KHP' which is reviewing diabetes services provided by the KHP institutions with a view to service re-design. Phase one of this project (data collection) is supported by Novo Nordisk. My current research is supported by NIHR and Diabetes UK. I am a co-investigator on a collaborative research grant from Medtronic.

Please see attached publication list. I am Chairman of the EFSD/China Diabetes Society/Lilly Programme, an EFSD body that awards research grants and fellowships and organises educational visits with CDS; I am Editor of the International Diabetes Federation’s journal Diabetes Voice; I participate in the National Director for Diabetes' insulin pump working party. I am on the Juvenile Diabetes Research Foundations' international scientific advisory board; academic member of the committee of the Association of British Clinical Diabetologist and a member of Diabetes UKs research committee.

GDG 2 No personal pecuniary interests within the last 12 None 27 November 2012 months. GDG 3 Non-personal pecuniary: A named investigator on a Declare and participate. 8 January 2013 research study being run by a colleague here at King's, Dr Pratik Choudhary, that is a collaboration with Medtronic, the insulin pump and glucose sensor manufacturer. The study is an investigation into the efficacy and safety of their overnight closed loop insulin delivery device and has been 3 years in coming to fruition, which is why it is now being activated. Dr Choudhary is the grant awardee. I did have input into the study design - again without pecuniary advantage - and will be involved in its delivery. The grant is made to KCL in Dr Choudhary's name and he is the Chief and Principal investigator, and no money comes to me. GDG 4 None declared None 5 March 2013 GDG 5 Serving on the international working party on None 9 April 2013 hypoglycaemia but there is no remuneration. GDG 6 None declared None 14 May 2013 GDG 7 None declared None 18 June 2013 GDG 8 Speaker at the SE/SW Diabetes Specialist Nurses Declare and participate 3 September 2013 forum on 22 November 2013. No remuneration.

National Clinical Guideline Centre, 2014 24 Type 1 diabetes in adults

Declarations of interest

GDG meeting Declaration of Interests Action taken GDG 9 None declared None 14 October 2013 GDG 10 None declared None 15 October 2013 GDG 11 None declared None 26 November 2013 GDG 12 None declared None 14 January 2014 GDG 13 None declared None 15 January 2014 GDG 14 None declared None 4 March 2014 GDG 15 None declared None 15 April 2014 GDG 16 Personal non-pecuniary: This European Association Declare and participate 20 May 2014 for the study of Diabetes (EASD) satellite symposium is being held by the International Hypoglycaemia Study Group (IHSG). It is funded and run by Six Degrees, which is funded by an unrestricted educational grant from Novo Nordisk. Six degrees is independent and has funding for different projects from many different sponsors. I accept travel costs and accommodation but no honorarium.

Personal non-pecuniary: I am giving the Jeff Goulder Memorial lecture at the Oxford Diabetes Symposium in June. The Oxford symposium is an annual event of high repute sponsored by Novo Nordisk but run independently by the Oxford Centre for Diabetes Endocrinology and Metabolism (OCDEM). I will receive travel expenses and accommodation in an Oxford College but no other pecuniary gain. GDG 17 None declared None 27 June 2014 GDG 18 Personal non-pecuniary interest: Participation Declare and participate. 29 July 2014 continuing in Six degrees hypoglycaemia study (as previously declared).

Non-personal pecuniary interest: Academic representative on King's Health Partners project with Novo Nordisk (as previously declared).

Personal non-pecuniary interest: will deliver a talk at the Diabetes Specialist Nurses forum on 6 November 2014. Non-promotional meeting. I will not receive any payment.

Personal non-pecuniary: gave permission and details for Kings College Hospital for the Accu-Chek DiaPort (Roche product - intraperitoneal insulin infusion port) brochure on centres of excellence for

National Clinical Guideline Centre, 2014 25 Type 1 diabetes in adults

Declarations of interest

GDG meeting Declaration of Interests Action taken healthcare professionals. GDG 19 Personal pecuniary interest: I attended the EASD Declare and participate 24 September 2014 (European Association for the Study of Diabetes) meeting which was paid for by six degrees (hypoglycaemia study group). Travel and accommodation expenses were received. GDG 20 TBC 24 March 2015

B.2 Arthur Durrant GDG meeting Declaration of Interests Action taken On appointment None declared None 8 August 2012 GDG 1 Lifelong membership of Diabetes UK. None 23 October 2012 GDG 2 None declared None 27 November 2012 GDG 3 None declared None 8 January 2013 GDG 4 None declared None 5 March 2013 GDG 5 None declared None 9 April 2013 GDG 6 None declared None 14 May 2013 GDG 7 None declared None 18 June 2013 GDG 8 None declared None 3 September 2013 GDG 9 Arthur declared that he has joined the lay advisory None 14 October 2013 panel for diabetes and endocrinology research at Sheffield Teaching Hospitals NHS Foundation Trust. GDG 10 None declared None 15 October 2013 GDG 11 None declared None 26 November 2013 GDG 12 None declared None 14 January 2014 GDG 13 None declared None 15 January 2014 GDG 14 None declared None 4 March 2014 GDG 15 None declared None 15 April 2014 GDG 16 I have recently been appointed as the chair of the None 20 May 2014 Sheffield Teaching Hospitals’ Lay Panel for Diabetes & Endocrinology Research.

National Clinical Guideline Centre, 2014 26 Type 1 diabetes in adults

Declarations of interest

GDG meeting Declaration of Interests Action taken GDG 17 None declared None 27 June 2014 GDG 18 None declared None 29 July 2014 GDG 19 TBC - 24 September 2014 GDG 20 TBC 24 March 2015

B.3 Michael Flynn GDG meeting Declaration of Interests Action taken On appointment None declared None 12 July 2012 GDG 1 Attended meetings supported by pharmaceutical Declare and participate 23 October 2012 companies. Reasonable expenses only. GDG 2 None declared None 27 November 2012 GDG 3 None declared None 8 January 2013 GDG 4 None declared None 5 March 2013 GDG 5 None declared None 9 April 2013 GDG 6 None declared None 14 May 2013 GDG 7 None declared None 18 June 2013 GDG 8 None declared None 3 September 2013 GDG 9 None declared None 14 October 2013 GDG 10 None declared None 15 October 2013 GDG 11 None declared None 26 November 2013 GDG 12 None declared None 14 January 2014 GDG 13 None declared None 15 January 2014 GDG 14 None declared None 4 March 2014 GDG 15 None declared None 15 April 2014 GDG 16 None declared None 20 May 2014

National Clinical Guideline Centre, 2014 27 Type 1 diabetes in adults

Declarations of interest

GDG meeting Declaration of Interests Action taken GDG 17 None declared None 27 June 2014 GDG 18 None declared None 29 July 2014 GDG 19 TBC - 24 September 2014 GDG 20 TBC 24 March 2015

B.4 Roger Gadsby GDG meeting Declaration of Interests Action taken On appointment My current industry work involves advisory boards Declare and withdraw from 29 January 2012 for MSD (sitagliptin), NovoNordisk (Insulin degludec) discussions of all types of and Janssen (Canagliflozin). Insulin degludec is the insulin. only one that I think has possible impacts on Type 1 diabetes.

Dr Roger Gadsby has received funding over the past 28 years for attending symposia and other educational events, for speaking at meetings, and for participating in advisory committees, from a variety of diabetes and cardiovascular pharmaceutical companies. These include AstraZeneca, Boehringer Ingelheim, Janssen-Cilag GlaxoSmithKline, Servier, Sanofi-Aventis, Takeda, Bristol Myers Squibb, NovoNordisk, Roche, Roche Diagnostics, MSD, Merck-Serono, Grunenthal, Solvay and Novartis.

He holds no shares in any companies.

From 1 Sept 2009 to 31 Dec 2011 two sessions per week of his time at Warwick Medical School were supported by an unrestricted educational grant from the two pharmaceutical companies NovoNordisk and Takeda.

Personal family interest: None

Non-personal pecuniary interest: In 2000 he helped to set up Warwick Diabetes Care (WDC), an organisation at Warwick University providing diabetes education programmes across the UK in 2000. WDC received support from 14 diabetes pharmaceutical companies who as foundation sponsors provided educational grants to assist its launch and initial development.

Several of the current diabetes education programmes at Warwick Medical School are supported by educational grants from pharmaceutical companies.

National Clinical Guideline Centre, 2014 28 Type 1 diabetes in adults

Declarations of interest

GDG meeting Declaration of Interests Action taken

He is chairman of the Trustees of a small charity called Pregnancy Sickness Support Trust (Registered Charity No 1094788) which gives information and support to women with pregnancy sickness symptoms. The charity has received support from a number of organisations and companies including Vitabiotics Ltd, Duchesnay Inc (Canada) and the charity committee of Land Rover plc. GDG 1 Not a member of the GDG None 23 October 2012 GDG 2 Not a member of the GDG None 27 November 2012 GDG 3 Not a member of the GDG None 8 January 2013 GDG 4 Did not attend None 5 March 2013 GDG 5 Doing advisory work paid for by Novo Nordisk Declare and withdraw from 9 April 2013 discussions on insulin. GDG 6 I was a member of the NovoNordisk Diabetes Declare and withdraw from 14 May 2013 Primary Care Advisory Board for Degludec which met discussions on insulin. on 23 October 2012. That is the only paid work that could possibly constitute a conflict of interest.

Other Advisory boards were for Janssen on 9 Nov 2012 and 2 May 2013 but these are for canagloflozin a drug for type 2 diabetes, so no conflict with the type 1 guideline I think.

My other strands of paid work in the past year are through my work at the Universities of Warwick and Bedfordshire, My work as GP lead for the National Diabetes Audit, and as Primary care lead for NHS Diabetes (finished on 31 March 2013). I have also done lots of work for NICE, most of it unpaid apart from 3 days paid as chairman of the Annual Evidence Update panel for CG 119. GDG 7 None declared None 18 June 2013 GDG 8 None declared None 3 September 2013 GDG 9 None declared None 14 October 2013 GDG 10 Did not attend meeting. None 15 October 2013 GDG 11 Did not attend meeting. None 26 November 2013 GDG 12 None declared None 14 January 2014 GDG 13 None declared None

National Clinical Guideline Centre, 2014 29 Type 1 diabetes in adults

Declarations of interest

GDG meeting Declaration of Interests Action taken 15 January 2014 GDG 14 None declared None 4 March 2014 GDG 15 Since the last GDG meeting I have been asked to Declare and participate. 15 April 2014 attend and have attended a first meeting of an advisory group for the Eli Lilly / Boehringer Alliance looking at a new insulin which is not licensed or launched yet in the UK. It is likely to come to market sometime in 2015. GDG 16 None declared None 20 May 2014 GDG 17 Did not attend meeting. None 27 June 2014 GDG 18 None declared None 29 July 2014 GDG 19 None declared 24 September 2014 GDG 20 TBC 24 March 2015

B.5 Peter Hammond GDG meeting Declaration of Interests Action taken On appointment Personal Pecuniary: In the last 12 months I have Declare and withdraw from 11 July 2012 received honoraria for lecturing from Novo Nordisk, discussing recommendations Lilly, Sanofi (ELIXA trial), Medtronic and Animas; and for insulin, continuous glucose for attending advisory boards from Sanofi (ELIXA monitoring and pumps (valid study) and Medtronic. until October 2013). GDG 1 Sanofi sponsored attendance at the EASD October Declare and withdraw from 23 October 2012 2012 (reasonable expenses). discussing recommendations for insulin, continuous glucose I lectured for both Animas and Medtronic in October monitoring and pumps (valid (16th and 24th respectively). I have future until October 2013). commitments lecturing for Medtronic but have advised them that I will no longer be paid for this work.

I have not been paid for lecturing this year by Novo, Sanofi or Lilly, so my last paid activity was an advisory board for Sanofi at Diabetes UK in March (8th I believe), although this was specifically an update meeting for researchers on the ELIXA study (lixisenatide - GLP1 agonist).

My unit is a centre for the ELIXA study. (Sanofi) NHS Diabetes Lead for Insulin Pump Therapy GDG 2 Personal Pecuniary: In the last 12 months I have Declare and withdraw from 27 November 2012 received honoraria for lecturing from Novo Nordisk, discussing recommendations Lilly, Roche, Medtronic and Animas; and for for insulin, continuous glucose attending advisory boards from Sanofi (ELIXA trial) monitoring and pumps (valid and Medtronic. Sanofi sponsored my attendance at until October 2013).

National Clinical Guideline Centre, 2014 30 Type 1 diabetes in adults

Declarations of interest

GDG meeting Declaration of Interests Action taken the EASD October 2012. GDG 3 None declared None 8 January 2013 GDG 4 Roche provided sponsorship to attend Advanced Declare and participate. 5 March 2013 Technology and Therapeutics in Diabetes meeting, Paris 27th February - 2nd March 2013. GDG 5 None declared None 9 April 2013 GDG 6 None declared None 14 May 2013 GDG 7 None declared None 18 June 2013 GDG 8 None declared None 3 September 2013 GDG 9 None declared None 14 October 2013 GDG 10 None declared None 15 October 2013 GDG 11 None declared None 26 November 2013 GDG 12 None declared None 14 January 2014 GDG 13 None declared None 15 January 2014 GDG 14 None declared None 4 March 2014 GDG 15 None declared None 15 April 2014 GDG 16 Personal non-pecuniary interest: attended an Declare and participate 20 May 2014 advisory panel for the flash glucose sensor. His attendance fee was donated to charity by Abbott. GDG 17 None declared None 27 June 2014 GDG 18 None declared None 29 July 2014 GDG 19 Personal non-pecuniary interest: I did a video- Declare and participate 24 September 2014 workshop on diabetes and pregnancy for Medtronic to various centres in Eastern Europe. GDG 20 TBC 24 March 2015

B.6 Michael Kendall GDG meeting Declaration of Interests Action taken On appointment None declared None 14 August 2012 GDG 1 Reasonable expenses received to attend a Declare and participate presentation by C8 Medisensors for new glucose

National Clinical Guideline Centre, 2014 31 Type 1 diabetes in adults

Declarations of interest

GDG meeting Declaration of Interests Action taken 23 October 2012 monitor (currently unavailable and in development) GDG 2 None declared None 27 November 2012 GDG 3 None declared None 8 January 2013 GDG 4 None declared None 5 March 2013 GDG 5 None declared None 9 April 2013 GDG 6 Acted as an unpaid / voluntary 'Alpha tester' Declare and participate 14 May 2013 between February 2013 and May 2013 for the new version (V2.0) of the mySugr diabetes logging app for iPhone http://mysugr.com which I have been using since September 2012. GDG 7 None declared None 18 June 2013 GDG 8 None declared None 3 September 2013 GDG 9 None declared None 14 October 2013 GDG 10 None declared None 15 October 2013 GDG 11 None declared None 26 November 2013 GDG 12 None declared None 14 January 2014 GDG 13 None declared None 15 January 2014 GDG 14 None declared None 4 March 2014 GDG 15 None declared None 15 April 2014 GDG 16 None declared None 20 May 2014 GDG 17 None declared None 27 June 2014 GDG 18 None declared None 29 July 2014 GDG 19 I was invited by Abbott to attend a pre-launch event Declare and withdraw from 24 September 2014 for the Libre Flash glucose monitoring system prior further discussions about to its launch at the European Association for the cotinuous glucose monitoring. Study of Diabetes meeting in September 2014. I was given a starter pack to allow a trial of the system for 28 days which has a retail value of £150. Reasonable travel expenses only were paid. GDG 20 TBC 24 March 2015

National Clinical Guideline Centre, 2014 32 Type 1 diabetes in adults

Declarations of interest

B.7 Vibhuti Mistry GDG meeting Declaration of Interests Action taken On appointment None declared None 20 July 2012 GDG 1 1. Sponsored by Animas to attend 2 day insulin None 23 October 2012 pump conference delivered by the Cambridge Diabetes Team. 2. Sponsored by Lilly to do an MSc module 'Insulin Management' at Leicester University. 3. 'Year of Care' Trainer (Care Planning part of NHS Diabetes). Teaching for Lilly and Novo over a year ago. GDG 2 None declared None 27 November 2012 GDG 3 None declared None 8 January 2013 GDG 4 None declared None 5 March 2013 GDG 5 Non-personal pecuniary: Vib has been asked by Lilly No action required for this 9 April 2013 to do some teaching for which she would not be meeting apoid directly (her department would revceive the payment). GDG 6 None declared None 14 May 2013 GDG 7 None declared None 18 June 2013 GDG 8 None declared None 3 September 2013 GDG 9 Did not attend meeting. None 14 October 2013 GDG 10 None declared None 15 October 2013 GDG 11 None declared None 26 November 2013 GDG 12 None declared None 14 January 2014 GDG 13 None declared None 15 January 2014 GDG 14 None declared None 4 March 2014 GDG 15 None declared None 15 April 2014 GDG 16 None declared None 20 May 2014 GDG 17 None declared None 27 June 2014 GDG 18 None declared None 29 July 2014

National Clinical Guideline Centre, 2014 33 Type 1 diabetes in adults

Declarations of interest

GDG meeting Declaration of Interests Action taken GDG 19 TBC - 24 September 2014 GDG 20 TBC 24 March 2015

B.8 Henrietta Mulnier GDG meeting Declaration of Interests Action taken On appointment Novo Nordisk - Nurse advisory board. One-off Declare and withdraw from 13 August 2012 consultancy. discussions about insulin and Lilly - occasional paid teaching sessions which may needles. continue.

GDG 1 Personal Pecuniary: In the past 12 months I have Declare and withdraw from 23 October 2012 received honoraria from Novo Nordisk, for attending discussions about insulin and a nurse advisory board, and from Lilly for teaching. I needles have also been paid a very modest fee for writing editorials and for the diabetes journals published by SB communications. GDG 2 Novo Nordisk, nurse advisory board. One-off Declare and withdraw from 27 November 2012 consultancy on 3rd July 2012. I will decline to attend discussions about insulin and any further boards needle lengths. Conflict for Occasional paid teaching for Lilly. Most recent and questions comparing Insulin last was on 2nd November 2012. I will decline future types (expires 2/11/2013) and teaching. needle length (expires Occasional writing for SB communications who 3/7/2013). Agreed to give up publish diabetes journals such as the Journal of paid work for industry for the Diabetes Nursing. I will write these without payment duration of the guideline. from now on. In summary In the past 12 months I have received honoraria from Novo Nordisk, for attending a nurse advisory board, and from Lilly for teaching. I have also been paid a very modest fee for writing editorials and for the diabetes journals published by SB communications. GDG 3 None declared None 8 January 2013 GDG 4 None declared None 5 March 2013 GDG 5 None declared None 9 April 2013 GDG 6 None. Did not attend meeting. None 14 May 2013 GDG 7 None declared None 18 June 2013 GDG 8 Personal non-pecuniary interest: invited to join a Declare and participate at 3 September 2013 working group funded by Novo Nordisk to develop a chair's discretion (pending new education and support programme. I will expiry of prior conflict for decline payment and accept only reasonable travel insulin). and accommodation expenses. GDG 9 None declared None

National Clinical Guideline Centre, 2014 34 Type 1 diabetes in adults

Declarations of interest

GDG meeting Declaration of Interests Action taken 14 October 2013 GDG 10 None declared None 15 October 2013 GDG 11 None declared None 26 November 2013 GDG 12 None declared None 14 January 2014 GDG 13 None declared None 15 January 2014 GDG 14 None declared None 4 March 2014 GDG 15 None declared None 15 April 2014 GDG 16 None declared None 20 May 2014 GDG 17 None declared None 27 June 2014 GDG 18 None declared None 29 July 2014 GDG 19 Non-personal pecuniary interest: I delivered a Declare and participate. 24 September 2014 module on behalf of Kings College London to East Lancashire Health Trust. The funding for the module came from Lilly; she did not receive a direct payment and carried out the teaching according to her role as Lecturer King’s College London. GDG 20 TBC 24 March 2015

B.9 Victoria Ruszala GDG meeting Declaration of Interests Action taken On appointment Personal Pecuniary: I have received consultancy fees Declare and withdraw from 13 August 2012 from the following companies for sponsored road discussions about insulin, shows and teaching sessions related to diabetes CGM and needles. (predominantly Type 2 diabetes): Lilly, Sanofi Aventis, Merck Sharpe Dohme, Novo Nordisk, Boehringer Ingelheim. UKCPA teaching at sponsored study days. GDG 1 None declared None 23 October 2012 GDG 2 UKCPA teaching at sponsored study days. Declare and withdraw from 27 November 2012 19/11/2012 Sanofi Free attendance at UKCPA discussions about insulin, Symposium CGM and needles. 16/02/2012 Pharmacy Management/ Novo Nordisk Steering Group (next meeting Jan 2013) 19/03/2012 Boehringer / Lilly Progression of Type 2 diabetes: Workshop 09/05/2012 MSD SMART update in Type 2 Diabetes: Workshop

National Clinical Guideline Centre, 2014 35 Type 1 diabetes in adults

Declarations of interest

GDG meeting Declaration of Interests Action taken 28/06/2012 Boehringer / Lily Addressing Kidney Health in Type 2 Diabetes: GP teaching 05/12/2012 Boehringer / Lily Complexity of Type 2 diabetes: Workshop (Last paid meeting)

GDG 3 None declared None 8 January 2013 GDG 4 Did not attend meeting None 5 March 2013 GDG 5 None declared None 9 April 2013 GDG 6 None declared None 14 May 2013 GDG 7 None declared None 18 June 2013 GDG 8 None declared None 3 September 2013 GDG 9 Personal non-pecuniary: Vicky is the pharmacy lead Declare and participate 14 October 2013 for NHS England Endocrinology clinical reference group. GDG 10 None. Did not attend meeting. None 15 October 2013 GDG 11 None. Did not attend meeting. None 26 November 2013 GDG 12 None declared None 14 January 2014 GDG 13 None declared None 15 January 2014 GDG 14 None declared None 4 March 2014 GDG 15 None declared None 15 April 2014 GDG 16 None declared None 20 May 2014 GDG 17 None declared None 27 June 2014 GDG 18 Personal pecuniary: Vicky Ruszala declared that she Declare and withdraw from 29 July 2014 has recently attended an advisory board for Sanofi, discussions on insulin. thus extending the active period for the interest already declared in relation to Sanofi. Vicky is conflicted for insulin and the chair agreed that she could not participate in reviewing these recommendations. GDG 19 None declared - 24 September 2014 GDG 20 TBC 24 March 2015

National Clinical Guideline Centre, 2014 36 Type 1 diabetes in adults

Declarations of interest

B.10 Stuart Smellie GDG meeting Declaration of Interests Action taken On appointment None None 13 July 2012 GDG 1 Occasional educational talks to GPs funded by a Declare and participate 23 October 2012 AstraZeneca, MSD, Schering Plough and Sanofi Aventis. Talks given twice a year, most years, about cholesterol lowering (standard rate of £300 per talk). MSD, Schering Plough and Sanofi were all more than 12 months ago. AstraZeneca was within the last 6 months.

Stuart was paid for attendingan advisory board for tredaptive (MSD), which predated the GDG.

Visiting professorship programme for Australian Royal College of Pathologists -travel and accommodation provided for visit in 2014.

Taught on BMJ Masterclasses - educational and paid but no industry involvement.

Clinical Director, Association of Clinical Biochemists. Vice President, Association of Clinical Pathologists.

GDG 2 Did not attend meeting. None 27 November 2012 GDG 3 None None 8 January 2013 GDG 4 None None 5 March 2013 GDG 5 None None 9 April 2013 GDG 6 None None 14 May 2013 GDG 7 Did not attend meeting. None 18 June 2013 GDG 8 None None 3 September 2013 GDG 9 None None 14 October 2013 GDG 10 None None 15 October 2013 GDG 11 None None 26 November 2013 GDG 12 Did not attend meeting. None 14 January 2014 GDG 13 Did not attend meeting. None 15 January 2014

National Clinical Guideline Centre, 2014 37 Type 1 diabetes in adults

Declarations of interest

GDG meeting Declaration of Interests Action taken GDG 14 None declared. None 4 March 2014 GDG 15 None declared. None 15 April 2014 GDG 16 None declared. None 20 May 2014 GDG 17 Did not attend meeting. None 27 June 2014 GDG 18 None declared. None 29 July 2014 GDG 19 TBC - 24 September 2014 GDG 20 TBC 24 March 2015

B.11 Perdy van den Berg GDG meeting Declaration of Interests Action taken On appointment None None 13 July 2012 GDG 1 Sponsorship from Novo Nordisk for £250 to cover No action required 23 October 2012 costs of travel, registration and overnight accommodation for PCOS annual conference in Birmingham 16-17th November. GDG 2 None None 27 November 2012 GDG 3 Will receive reasonable expenses to attend Diabetes No action required 8 January 2013 UK conference GDG 4 None declared None 5 March 2013 GDG 5 Did not attend meeting. None 9 April 2013 GDG 6 None declared None 14 May 2013 GDG 7 None declared None 18 June 2013 GDG 8 None declared None 3 September 2013 GDG 9 Did not attend meeting. None 14 October 2013 GDG 10 None declared None 15 October 2013 GDG 11 Did not attend meeting. None 26 November 2013 GDG 12 None declared None 14 January 2014 GDG 13 None declared None

National Clinical Guideline Centre, 2014 38 Type 1 diabetes in adults

Declarations of interest

GDG meeting Declaration of Interests Action taken 15 January 2014 GDG 14 None declared None 4 March 2014 GDG 15 None declared None 15 April 2014 GDG 16 Did not attend meeting. None 20 May 2014 GDG 17 None declared None 27 June 2014 GDG 18 None declared None 29 July 2014 GDG 19 TBC - 24 September 2014 GDG 20 TBC 24 March 2015

B.12 Rayaz Malik (co-opted expert) GDG meeting Declaration of Interests Action taken On appointment None declared No action required 13 July 2012 GDG 12 None declared No action required 14 January 2014 GDG 16 Non declared No action required 20 May 2014

B.13 NCGC members GDG meeting Declaration of Interests Action taken On appointment Bernard Higgins: Non-personal pecuniary interest: Declare and participate 13 July 2012 The department in which I work has taken part in multi-centre studies over the past 12 months, sponsored by the following companies: Novartis,

Pro-pharma, Schering-Plough, Aradigm, GSK,

ResMed, Astra-Zeneca, Nycomed, Gilead and Chiesi. I do not receive any personal payment or gratuity, and I do not have administrative responsibility for the fund into which payments are made. Personal non-pecuniary interest: I am the chair-elect Declare and participate of BTS executive.

Elisabetta Fenu: Personal pecuniary interest: I was Declare and participate an employee of Novartis UK where I held the position of Health Econoics and Outcomes Manager

in May and June 2012. I was involved in the NICE TA submission of the product Xolair for the treatment Declare and participate of asthma.

Nancy Turnball: Personal pecuniary interest: I do consultancy work for WHO and other departments

National Clinical Guideline Centre, 2014 39 Type 1 diabetes in adults

Declarations of interest

GDG meeting Declaration of Interests Action taken of the RCP. GDG 1 None declared No action required 23 October 2012 GDG 2 None declared No action required 27 November 2012 GDG 3 None declared No action required 8 January 2013 GDG 4 None declared No action required 5 March 2013 GDG 5 None declared No action required 9 April 2013 GDG 6 None declared No action required 14 May 2013 GDG 7 Non e declared No action required 18 June 2013 GDG 8 None declared No action required 3 September 2013 GDG 9 None declared No action required 14 October 2013 GDG 10 None declared No action required 15 October 2013 GDG 11 None declared No action required 26 November 2013 GDG 12 None declared No action required 14 January 2014 GDG 13 None declared No action required 15 January 2014 GDG 14 None declared No action required 4 March 2014 GDG 15 None declared No action required 15 April 2014 GDG 16 None declared No action required 20 May 2014 GDG 17 None declared No action required 27 June 2014 GDG 18 None declared No action required 29 July 2014 GDG 19 None declared No action required 24 September 2014 GDG 20 TBC 24 March 2015

National Clinical Guideline Centre, 2014 40 Type 1 diabetes in adults

Review protocols

Appendix C: Review protocols

C.1 Clinical protocols

C.1.1 Diagnosis/markers for distinguishing types of diabetes

C.1.1.1 Distinguishing between different types of diabetes Component Description Review question In adults with diabetes, what is the best marker (c-peptides plus or minus antibodies) to distinguish between a diagnosis of type 1 diabetes, type 2 diabetes and other forms of diabetes? Objectives The aim of this review is to determine whether the presence of c-peptides and/or antibodies in people with diabetes distinguishes between type 1 diabetes, type 2 diabetes, and other forms of diabetes, in order to make an accurate diagnosis. Also what titre/concentration of each of these markers is present and distinguishes the types.  What is the best test or combination of tests to give you the highest level of certainty in discriminating type 1 diabetes from type 2 diabetes?  When is there uncertainty, and what should be done when uncertainty exists? Population Adults with all types of diabetes  Adult is defined as aged ≥18 years  Diabetes types are: type 1 diabetes, type 2 diabetes, LADA and MODY. Subgroups The following groups will be considered separately if data are available:  Adults Diagnostic tests  C-peptide o Plasma C-peptide (stimulated) o Urinary C-peptide o Urinary C-peptide/creatinine ratio

 Antibody tests: o anti-islet cell antibody (ICA) o anti-glutamic acid decarboxylase 65 antibody or anti-glutamic acid decarboxylase antibody (GADA) o insulinoma-associated (IA-2/ICA512) autoantibody o zinc transporter 8 (ZnT8) o islet-specific glucose-6-phosphatase catalytic subunit (IGRP) Outcomes  Presence of marker (No. or % of patients with marker)  Concentration/titre of marker  Change in marker over time (No. or % of patients with marker)  Change in concentration/titre of marker over time (micrograms/ml) Importance of Critical outcomes outcomes  N/A Study design  Observational studies Population size  Studies will be excluded if: and directness o The population is mixed, with no adult subgroup analyses. Mixed populations excluded are: - Children and young people

National Clinical Guideline Centre, 2014 41 Type 1 diabetes in adults

Review protocols

- Children, young people and adults o The population is exclusively children (age <11 years) o they are validation studies o they are treatment studies o they are pre-diabetes populations (we are not going to look at studies that use markers as predictors of the future development of diabetes) o they are detecting markers in relatives of people with diabetes  Sample size restrictions. We will exclude studies in: o Adults and young people with sample size of N<50, if we retrieve >20 studies that have been conducted in adults and young people separately o Adults with a sample size of N<50, if there are >20 adult studies retrieved. Setting  All settings (as per Scope) Search Strategy See appendix F Search will be restricted to studies published since the original GL (2003). If only a few studies are found, then we will extend the search to all years. Review Strategy Appraisal of methodological quality/evidence synthesis  The methodological quality of each study will be assessed using NICE checklists and a narrative synthesis of the evidence will be provided. Notes If no/insufficient evidence is found we will (in order of preference):  Consider unpublished or partially published studies (including abstracts – and contact the authors for more information)  Move to GDG consensus

C.1.2 Educating programmes and self-care

C.1.2.1 Structured education programmes Component Description Review question In adults with type 1 diabetes, what is the most effective structured education programme in terms of clinical and cost-effectiveness? Objectives The aim of this review is to compare different structured educational programmes for adults with type 1 diabetes and whether they lead to an improvement in QoL and clinical outcomes Population Adults with type 1 diabetes  Adult is defined as aged over 18 years Subgroups The following groups will be considered separately if data are available:  None specified

The following factors will be considered for subgroup analysis if heterogeneity is present, in the following order:  Whether patients included or recruited in the trials were solely type 1 diabetes with hypoglycaemia  Duration of diabetes (for example newly diagnosed)  Whether CHO restriction was included in the programme

If heterogeneity is still not explained then we will look at:  Whether follow-up programmes were done or not  Basal insulin regimens: o once versus twice/day o analogue versus non-analogue

National Clinical Guideline Centre, 2014 42 Type 1 diabetes in adults

Review protocols

Intervention  Structured education programme Comparison  Other education programmes  Usual care/no treatment  SMBG Outcomes  HbA1c (continuous)  Hypoglycaemia - preferably severe hypoglycaemia if reported (dichotomous or continuous outcome, depending how it is reported)  Hospital admissions (dichotomous or continuous, depending how it is reported)  Hypoglycaemia unawareness (dichotomous)  Quality of life – measured by DQoL, DSQoL, PAID, HADS, fear of hypoglycaemia, anxiety, depression (continuous)  Adverse events (dichotomous)  Knowledge (dichotomous or continuous, depending how it is reported)  Adherence (dichotomous or continuous, depending how it is reported) Importance of Critical outcomes outcomes  HbA1c  Severe hypoglycaemia  Quality of life Study design RCTs  Unit of randomisation: individual patient, cluster randomised trials Population size  No restrictions on sample size and directness  No restrictions on treatment duration  Studies with indirect, or mixed diabetes (type 1 diabetes and type 2 diabetes) populations will not be considered unless ≥75% are type 1 diabetes, or data has been reported for the subgroup of type 1 diabetes patients, in which case this subgroup data will be used. Setting  All settings (as per Scope) Search Strategy See appendix F Review Strategy Appraisal of methodological quality  The methodological quality of each study will be assessed using NICE checklists and the quality of the evidence will be assessed by GRADE for each outcome. Synthesis of data  Meta-analysis will be conducted where appropriate  Outcomes will be grouped into the following categories based on time-points: o ≤6 months (or the one nearest to 6 months if multiple time-points are given) o >6 months (or the longest one if multiple time-points are given)  Default MIDs will be used: 0.75 and 1.25 for dichotomous outcomes; 0.5 times SD for continuous outcomes Notes If no/insufficient RCT evidence is found we will (in order of preference):  Consider evidence from non-randomised, non-comparative and observational studies  Consider unpublished or partially published studies (including abstracts – and contact the authors for more information)  Move to GDG consensus

C.1.2.2 Carb counting In adults with type 1 diabetes, what is the clinical and cost-effectiveness of a diet Review question based on carbohydrate counting/restriction for optimal diabetic control? Objectives The aim of this review is to determine whether carbohydrate counting/restriction is effective for optimal diabetic blood glucose control in adults with type 1 diabetes in terms of cost and clinical effectiveness, particularly for post-prandial blood glucose control.

National Clinical Guideline Centre, 2014 43 Type 1 diabetes in adults

Review protocols

Population Adults with type 1 diabetes  Adult is defined as aged ≥ 18 years Subgroups The following subgroups will be considered separately if data are available:  None specified The following factors will be considered for subgroup analysis (in the critical outcomes only) if heterogeneity is present:  different methods of CHO counting/restriction  different thresholds set for CHO counting/restriction Intervention Carbohydrate counting /restriction (this may involve technology such as a bolus calculator)

Comparison  Placebo  Usual care/no carbohydrate counting  Manual carbohydrate counting (if the intervention is carb counting using a technology)

Outcomes  HbA1c (continuous)  Hypoglycaemia (dichotomous or continuous outcome, depending how it is reported)  Severe hypoglycaemia (dichotomous or continuous outcome, depending how it is reported)  Nocturnal hypoglycaemia (dichotomous or continuous outcome, depending how it is reported)  Quality of life – measured by whatever is used in the study (continuous)  Adverse events (dichotomous) Importance of Critical outcomes outcomes  HbA1c  Hypoglycaemia Study Design RCTs, observational studies  Unit of randomisation: individual patient Population size  No restrictions on sample size and directness  No restrictions on treatment duration  Studies with indirect, or mixed diabetes (type 1 diabetes and type 2 diabetes) populations will only be considered, if: o data has been reported for the subgroup of type 1 diabetes patients, in which case this subgroup data will be used, o the population contains ≥70% of type 1 diabetes patients, o if ≥50% of people are aged >18 years the study will be included.  Studies looking at different carbohydrate content diets will be excluded.  Studies looking at GI index will be excluded (as this is being covered in a separate review). Setting  All settings (as per Scope) Search Strategy See appendix F Review strategy Appraisal of methodological quality  The methodological quality of each study will be assessed using NICE checklists  Where possible, the quality of the evidence will be assessed by GRADE for each outcome.  Taking into consideration the advice on prognostic reviews in the NICE guidelines manual, meta-analysis or GRADE will not be undertaken for prognostic studies or qualitative studies

National Clinical Guideline Centre, 2014 44 Type 1 diabetes in adults

Review protocols

Synthesis of data  Meta-analysis will be conducted where appropriate  Outcomes will be grouped into the following categories based on time-points: o ≤6 months (or the one nearest to 6 months if multiple time-points are given) o >6 months (or the longest one if multiple time-points are given)  Default MIDs will be used: 0.75 and 1.25 for dichotomous outcomes; 0.5 times SD for continuous outcomes Notes If no/insufficient RCT evidence is found we will (in order of preference):  Consider evidence from non-randomised, non-comparative and observational studies  Consider unpublished or partially published studies (including abstracts – and contact the authors for more information)  Move to GDG consensus

C.1.2.3 Glycaemic index diet – low GI versus high GI diet Component Description Review question In adults with type 1 diabetes, what is the clinical and cost-effectiveness of a diet based on the glycaemic index for optimal diabetic control? Objectives The aim of this review is to determine whether a low GI diet is more effective than a high GI diet for diabetic control in adults with type 1 diabetes. Population Adults with type 1 diabetes  Adult is defined as aged ≥18 years Subgroups The following groups will be considered separately if data are available:  None specified

The following factors will be considered for subgroup analysis (in the critical outcomes only) if heterogeneity is present:  different background insulin regimens Intervention High GI diet (this may include balancing protein and lipid intake) Comparison Low GI diet (this may include balancing protein and lipid intake) Outcomes  HbA1c (continuous)  Severe hypoglycaemia (dichotomous or continuous outcome, depending how it is reported)  Nocturnal hypoglycaemia (dichotomous or continuous outcome, depending how it is reported)  Quality of life – measured by DQoL or any measure used in the studies retrieved (continuous)  Patient satisfaction (dichotomous or continuous outcome, depending how it is reported)  Adherence (dichotomous) Importance of Critical outcomes outcomes  HbA1c  Severe hypoglycaemia Study design RCTs  Unit of randomisation: individual patient Population size and  No restrictions on sample size directness  No restrictions on treatment duration  Studies with indirect, or mixed diabetes (type 1 diabetes and type 2 diabetes) populations will only be considered, if: o data has been reported for the subgroup of type 1 diabetes patients, in which case this subgroup data will be used.

National Clinical Guideline Centre, 2014 45 Type 1 diabetes in adults

Review protocols

o the population contains ≥70% of type 1 diabetes patients o if ≥50% of people are aged >18 years the study will be included Setting  All settings (as per Scope) Search Strategy See appendix F Review Strategy Appraisal of methodological quality  The methodological quality of each study will be assessed using NICE checklists and the quality of the evidence will be assessed by GRADE for each outcome. Synthesis of data  Meta-analysis will be conducted where appropriate  Outcomes will be grouped into the following categories based on time-points: o ≤6 months (or the one nearest to 6 months if multiple time-points are given) o >6 months (or the longest one if multiple time-points are given)  Default MIDs will be used: 0.75 and 1.25 for dichotomous outcomes; 0.5 times SD for continuous outcomes Notes If no/insufficient RCT evidence is found we will (in order of preference):  Consider evidence from non-randomised, non-comparative and observational studies  Consider unpublished or partially published studies (including abstracts – and contact the authors for more information)  Move to GDG consensus

C.1.3 Blood glucose monitoring

C.1.3.1 HbA1c (targets) Component Description Review question In adults with type 1 diabetes, what is the optimum target HbA1c level that should be achieved to reduce the risk of complications?

Objectives The aim of this review is to determine whether the current HbA1c target for adults with type 1 diabetes needs changing, as there may be evidence that lower targets lead to better diabetic control or outcomes, however there may be issues of adherence. Also to determine the risk of complications at different levels of HbA1c, and what is the optimum treatment regime that should follow to achieve this? Different targets for different groups of patients will be needed (for example, for pump users). Population Adults with type 1 diabetes  Adult is defined as aged ≥ 18 years Subgroups None

Intervention HbA1c target values Comparison  Other target values (RCTs and comparative observational studies)  No targets (prognostic studies) Outcomes  Number of people reaching target HbA1c (dichotomous)  Final HbA1cvalue (continuous)  Hypoglycaemia (dichotomous or continuous outcome at a particular target  Severe hypoglycaemia (dichotomous or continuous outcome, depending how it is reported)  Nocturnal hypoglycaemia (dichotomous or continuous outcome, depending how it is reported)  Complications/avoidance: o CV events (MI, IHD, Stroke, cardiac and peripheral revascularisation, major amputation)

National Clinical Guideline Centre, 2014 46 Type 1 diabetes in adults

Review protocols

Component Description o Retinopathy o Low-level (micro) albuminuria/proteinuria o Renal replacement therapy/end-stage renal failure o Neuropathy o Sudden death  Quality of life – (dichotomous/continuous) Importance of Critical outcomes outcomes  Complications  Hypoglycaemia Study design RCTs, observational studies, prognostic studies Population size  No restrictions on sample size (unless >10 studies, then sample sizes of N<100 will be and directness excluded)  No restrictions on study duration  Studies with indirect, or mixed diabetes (type 1 diabetes and type 2 diabetes ) populations will only be considered, if: o data has been reported for the subgroup of type 1 diabetes patients, in which case this subgroup data will be used. o the population contains ≥70% of type 1 diabetes patients o if ≥ 50% of people are aged >18 years the study will be included  Studies will be excluded if they: o do not stratify results by different HbA1ctargets Setting  All settings (as per Scope) Search Strategy See appendix F Review Strategy Appraisal of methodological quality  The methodological quality of each study will be assessed using NICE checklists  Where possible, the quality of the evidence will be assessed by GRADE for each outcome.  Taking into consideration the advice on prognostic reviews in the NICE guidelines manual, meta-analysis or GRADE will not be undertaken for prognostic studies Synthesis of data  Meta-analysis will be conducted where appropriate  Data will be synthesised in a narrative review where meta-analysis is not possible  Outcomes will be grouped into the following categories based on time-points:  ≤6 months (or the one nearest to 6 months if multiple time-points are given)  >6 months (or the longest one if multiple time-points are given)  Default MIDs will be used: 0.75 and 1.25 for dichotomous outcomes; 0.5 times SD for continuous outcomes Notes If no/insufficient RCT, observational or prognostic evidence is found, we will (in order of preference):  Consider unpublished or partially published studies (including abstracts – and contact the authors for more information)  Move to GDG consensus

C.1.3.2 HbA1c (frequency of monitoring) Component Description Review question In adults with type 1 diabetes, what is optimum frequency of HbA1c monitoring for effective diabetic control? Objectives The aim of this review is to determine how often HbA1c should be monitored in order to maintain optimum glucose/diabetic control and reduce the incidence of adverse events and hypoglycaemia episodes.

National Clinical Guideline Centre, 2014 47 Type 1 diabetes in adults

Review protocols

Component Description Population Adults with type 1 diabetes  Adult is defined as aged ≥ 18 years Subgroups The following groups will be considered separately if data are available:  Newly diagnosed (<5 years)  Significant microvascular complications  Hypoglycaemia unawareness The following factors will be considered for subgroup analysis (in the critical outcomes only) if heterogeneity is present:  None specified Intervention HbA1c monitoring Comparison  HbA1c monitoring (the same as the intervention but at a different frequency or delivery time)  Standard care  No comparison (non-comparative studies) Outcomes  Hypoglycaemia - preferably severe hypoglycaemia if reported (dichotomous or continuous outcome, depending how it is reported)  HbA1c (continuous)  Quality of life – measured by any measure reported in the study (continuous)  Adverse events (dichotomous)  Adherence (dichotomous)  Complications o CV events (MI, IHD, Stroke, cardiac and peripheral revascularisation, major amputation) o Retinopathy o Low-level (micro) albuminuria/proteinuria o Renal replacement therapy/end-stage renal failure o Neuropathy o Sudden death Importance of Critical outcomes outcomes  Hypoglycaemia  HbA1c Complications- retinopathy Study design All types Population size  No restrictions on sample size (unless >10 studies, then sample sizes of N<100 will be and directness excluded)  No restrictions on study duration  Studies with indirect, or mixed diabetes (type 1 diabetes and type 2 diabetes) populations will only be considered, if: o data has been reported for the subgroup of type 1 diabetes patients, in which case this subgroup data will be used. o the population contains ≥70% of type 1 diabetes patients o if ≥50% of people are aged >18 years the study will be included  Studies will be excluded if they: o do not stratify results by different monitoring frequencies Setting  All settings (as per Scope) Search Strategy See appendix F Review Strategy Appraisal of methodological quality

National Clinical Guideline Centre, 2014 48 Type 1 diabetes in adults

Review protocols

Component Description  The methodological quality of each study will be assessed using NICE checklists  Where possible, the quality of the evidence will be assessed by GRADE for each outcome.  Taking into consideration the advice on prognostic reviews in the NICE guidelines manual, meta-analysis or GRADE will not be undertaken for prognostic studies Synthesis of data  Meta-analysis will be conducted where appropriate  Data will be synthesised in a narrative review where meta-analysis is not possible  Outcomes will be grouped into the following categories based on time-points:  ≤6 months (or the one nearest to 6 months if multiple time-points are given)  >6 months (or the longest one if multiple time-points are given)  Default MIDs will be used: 0.75 and 1.25 for dichotomous outcomes; 0.5 times SD for continuous outcomes Notes If no/insufficient RCT, observational or prognostic evidence is found, we will (in order of preference):  Consider unpublished or partially published studies (including abstracts – and contact the authors for more information)  Move to GDG consensus

C.1.3.3 SMBG (timing and frequency) Component Description Review question In adults with type 1 diabetes, what is optimum timing and frequency to self-monitor blood glucose for effective diabetic control? Objectives The aim of this review is to determine how often SMBG should be performed in order to maintain optimum glucose/diabetic control and reduce the incidence of adverse events and hypoglycaemia episodes. Issues of adherence too? Population Adults with type 1 diabetes  Adult is defined as aged ≥ 18 years Subgroups The following groups will be considered separately if data are available:  significant comorbidities  hypoglycaemia problems  people with recurrent DKA  different durations of disease  different lifestyles/activity levels  people striving for tight control

The following factors will be considered for subgroup analysis (in the critical outcomes only) if heterogeneity is present:  different sites of monitoring  different times of monitoring  different technologies of monitoring  different glucose targets for monitoring Intervention SMBG (finger pricks) Comparison  SMBG (finger pricks) - the same as the intervention but at a different frequency or delivery time  No comparison (for non-comparative studies) Outcomes  Hypoglycaemia (dichotomous or continuous outcome, depending how it is reported)  Severe hypoglycaemia (dichotomous or continuous outcome, depending how it is

National Clinical Guideline Centre, 2014 49 Type 1 diabetes in adults

Review protocols

reported)  Nocturnal hypoglycaemia (dichotomous or continuous outcome, depending how it is reported)  Time within range (blood glucose) (continuous)  HbA1c (continuous)  Quality of life – measured by any measure specified in the study (continuous)  DKA (dichotomous)  Adherence (dichotomous)  Unscheduled care use Importance of Critical outcomes outcomes  Hypoglycaemia  HbA1c Study design  RCTs, observational studies any Population size  No restrictions on sample size (unless >10 studies, then sample sizes of N<100 will be and directness excluded)  No restrictions on study duration  Studies with indirect, or mixed diabetes (type 1 diabetes and type 2 diabetes) populations will only be considered, if: o data has been reported for the subgroup of type 1 diabetes patients, in which case this subgroup data will be used. o the population contains ≥70% of type 1 diabetes patients o if ≥50% of people are aged >18 years the study will be included  Studies will be excluded if they: o do not stratify results by different monitoring frequencies Setting  All settings (as per Scope) Search Strategy See appendix F Review Strategy Appraisal of methodological quality  The methodological quality of each study will be assessed using NICE checklists  Where possible, the quality of the evidence will be assessed by GRADE for each outcome.  Taking into consideration the advice on prognostic reviews in the NICE guidelines manual, meta-analysis or GRADE will not be undertaken for prognostic studies Synthesis of data  Meta-analysis will be conducted where appropriate  Data will be synthesised in a narrative review where meta-analysis is not possible  Outcomes will be grouped into the following categories based on time-points: o ≤6 months (or the one nearest to 6 months if multiple time-points are given) o >6 months (or the longest one if multiple time-points are given)  Default MIDs will be used: 0.75 and 1.25 for dichotomous outcomes; 0.5 times SD for continuous outcomes Notes If no/insufficient RCT, observational or prognostic evidence is found, we will (in order of preference):  Consider unpublished or partially published studies (including abstracts – and contact the authors for more information)  Move to GDG consensus

C.1.3.4 SMBG (targets) Component Description Review question In adults with type 1 diabetes, what is the optimum glucose target/profile for self-

National Clinical Guideline Centre, 2014 50 Type 1 diabetes in adults

Review protocols

monitoring of blood glucose for effective diabetic control? Objectives The aim of this review is to determine what is the best glucose target patients should be aiming for when self-monitoring blood glucose in order to lead to optimum diabetic control, however, there may be issues of adherence. We are interested in the impact of variability on complication risk. There is some data on variability. Do clinicians and patients need to worry about post prandial glucose levels if HbA1c is OK? Does HbA1c variability for complications matter? And what extent does mean blood glucose mean anything at all, given the HbA1c values. Population Adults with type 1 diabetes  Adult is defined as aged ≥18 years Subgroups The following groups will be considered separately if data are available:  Male versus female  Age The following factors will be considered for subgroup analysis (in the critical outcomes only) if heterogeneity is present:  Duration of diabetes  Presence of complications (do we need more monitoring if tighter control is needed?) Intervention SMBG (finger pricks) - Blood glucose target/profile values/glucose variabiity Comparison  Other target values (RCTs and comparative observational studies)  No targets (prognostic studies) Outcomes  HbA1c value (continuous)  Risk of hypoglycaemia (dichotomous or continuous)  Risk of severe hypoglycaemia (dichotomous or continuous)  Risk of nocturnal hypoglycaemia (dichotomous or continuous)  Risk of complications (dichotomous or continuous)  ‘QoL - any measure reported in the study (continuous) Importance of Critical outcomes outcomes  Risk of hypoglycaemia  Risk of complications Study design  RCTs, observational studies, prognostic studies Population size  No restrictions on sample size (unless >10 studies, then sample sizes of N<100 will be and directness excluded)  No restrictions on study duration  Studies with indirect, or mixed diabetes (type 1 diabetes and type 2 diabetes) populations will only be considered, if: o data has been reported for the subgroup of type 1 diabetes patients, in which case this subgroup data will be used. o the population contains ≥70% of type 1 diabetes patients o if ≥50% of people are aged >18 years the study will be included

 Studies will be excluded if they: o do not stratify results by different glucose targets Setting  All settings (as per Scope) Search Strategy See appendix F Review Strategy Appraisal of methodological quality  The methodological quality of each study will be assessed using NICE checklists  Where possible , the quality of the evidence will be assessed by GRADE for each outcome.  Taking into consideration the advice on prognostic reviews in the NICE guidelines

National Clinical Guideline Centre, 2014 51 Type 1 diabetes in adults

Review protocols

manual, meta-analysis or GRADE will not be undertaken for prognostic studies Synthesis of data  Meta-analysis will be conducted where appropriate  Data will be synthesised in a narrative review where meta-analysis is not possible  Outcomes will be grouped into the following categories based on time-points: o ≤6 months (or the one nearest to 6 months if multiple time-points are given) o >6 months (or the longest one if multiple time-points are given)  Default MIDs will be used: 0.75 and 1.25 for dichotomous outcomes; 0.5 times SD for continuous outcomes Notes If no/insufficient RCT, observational or prognostic evidence is found, we will (in order of preference):  Consider unpublished or partially published studies (including abstracts – and contact the authors for more information)  Move to GDG consensus

C.1.3.5 SMBG technologies Component Description Review question In adults with type 1 diabetes, what are the benefits of technologies (bolus calculators and downloads) for self-monitoring of blood glucose? Objectives The aim of this review is to determine whether new technologies (bolus calculators and downloads) improve SMBG and clinical outcomes. Issues of adherence too? Population Adults with type 1 diabetes  Adult is defined as aged ≥ 18 years Subgroups The following groups will be considered separately if data are available:  Age  Duration of diabetes  Blood glucose control (HbA1c <8% versus >8%)

The following factors will be considered for subgroup analysis (in the critical outcomes only) if heterogeneity is present:  different frequencies of monitoring  different site of monitoring  different times of monitoring  different glucose targets for monitoring Intervention  SMBG (finger pricks) - bolus calculators  SMBG (finger pricks) - downloads Comparison SMBG (finger pricks) – standard SMBG methods Outcomes  Hypoglycaemia (dichotomous or continuous outcome, depending how it is reported)  Severe hypoglycaemia (dichotomous or continuous outcome, depending how it is reported)  Nocturnal hypoglycaemia (dichotomous or continuous outcome, depending how it is reported)  HbA1c (continuous)  Quality of life – measured by xxxxxx (continuous)  Adverse events (dichotomous)  Adherence (dichotomous) Importance of Critical outcomes outcomes  Hypoglycaemia  HbA1c

National Clinical Guideline Centre, 2014 52 Type 1 diabetes in adults

Review protocols

Component Description Study design RCTs Population size  No restrictions on sample size (unless >10 studies, then sample sizes of N<100 will be and directness excluded)  No restrictions on treatment duration.  Studies with indirect, or mixed diabetes (type 1 diabetes and type 2 diabetes) populations will only be considered, if: o Data has been reported for the subgroup of type 1 diabetes patients, in which case this subgroup data will be used. o the population contains ≥70% of type 1 diabetes patients o if ≥50% of people are aged >18 years the study will be included Setting  All settings (as per Scope) Search Strategy See appendix F Review Strategy Appraisal of methodological quality  The methodological quality of each study will be assessed using NICE checklists  Where possible, the quality of the evidence will be assessed by GRADE for each outcome. Synthesis of data  Meta-analysis will be conducted where appropriate  Data will be synthesised in a narrative review where meta-analysis is not possible  Outcomes will be grouped into the following categories based on time-points: o ≤6 months (or the one nearest to 6 months if multiple time-points are given) o >6 months (or the longest one if multiple time-points are given)  Default MIDs will be used: 0.75 and 1.25 for dichotomous outcomes; 0.5 times SD for continuous outcomes Notes If no/insufficient RCT, observational or prognostic evidence is found, we will (in order of preference):  Observational studies  Consider unpublished or partially published studies (including abstracts – and contact the authors for more information)  Move to GDG consensus

C.1.3.6 CGM (retrospective) versus SMBG Component Description Review question In adults with type 1 diabetes, is retrospective CGM more effective than care without CGM (with SMBG) for improving diabetic control? Objectives The aim of this review is to determine whether retrospective CGM is more effective than no CGM (that is, SMBG) for managing diabetic control in adults with type 1 diabetes. Adherence may also be an issue? Population Adults with type 1 diabetes  Adult is defined as aged ≥ 18 years Subgroups The following groups will be considered separately if data are available:  None specified

The following factors will be considered for subgroup analysis if heterogeneity is present:  frequency of SMBG  Type of CGM or SMBG method/monitor Intervention Retrospective CGM

National Clinical Guideline Centre, 2014 53 Type 1 diabetes in adults

Review protocols

Comparison Care without CGM (with SMBG) – comparison group must be on similar insulin regimen as the treatment group Outcomes Outcomes  HbA1c (continuous)  Hypoglycaemia - preferably severe hypoglycaemia if reported (dichotomous or continuous outcome, depending how it is reported)  Quality of life – measured by any measure stated in paper (continuous) or patient satisfaction  Adverse events – (dichotomous)  Adherence (dichotomous) Importance of Critical outcomes outcomes  HbA1c  Hypoglycaemia Study design RCTs  Unit of randomisation: individual patient Population size  No restrictions on sample size and directness  Exclude studies <1 week duration  Studies with indirect, or mixed diabetes (type 1 diabetes and type 2 diabetes) populations will only be considered, if: o data has been reported for the subgroup of type 1 diabetes patients, in which case this subgroup data will be used, o the population contains ≥70% of type 1 diabetes patients, o if ≥50% of people are aged >18 years the study will be included.

Setting  All settings (as per Scope) Search Strategy See appendix F Review Strategy Appraisal of methodological quality  The methodological quality of each outcome will be assessed using NICE checklists and the quality of the evidence will be assessed by GRADE for each outcome. Synthesis of data  Meta-analysis will be conducted where appropriate  Outcomes will be grouped into the following categories based on time-points: o ≤6 months (or the one nearest to 6 months if multiple time-points are given) o >6 months (or the longest one if multiple time-points are given)  Default MIDs will be used to assess imprecision: 0.75 and 1.25 for dichotomous outcomes; 0.5 times SD for continuous outcomes Notes If no/insufficient RCT evidence is found we will (in order of preference):  Consider evidence from non-randomised, non-comparative and observational studies  Consider unpublished or partially published studies (including abstracts – and contact the authors for more information)  Move to GDG consensus

C.1.3.7 CGM (real-time) versus SMBG Component Description Review question In adults with type 1 diabetes, is real-time CGM more effective than SMBG CGM for optimum diabetic control? Objectives The aim of this review is to determine whether real-time CGM is more effective than care without CGM (that is, SMBG) for managing diabetic control in adults with type 1 diabetes. Adherence may also be an issue? Population Adults with type 1 diabetes

National Clinical Guideline Centre, 2014 54 Type 1 diabetes in adults

Review protocols

 Adult is defined as aged ≥18 years Subgroups The following groups will be considered separately if data are available:  None specified

The following factors will be considered for subgroup analysis if heterogeneity is present:  frequency of CGM (real-time)  Type of CGM method/monitor (real-time and retrospective)  Frequency of SMBG Intervention Real-time CGM Comparison Care without CGM (with SMBG) – comparison group must be on similar insulin regimen as the treatment group Outcomes  HbA1c (continuous)  Hypoglycaemia - preferably severe hypoglycaemia if reported (dichotomous or continuous outcome, depending how it is reported)  Quality of life – measured by any measure stated in the paper (continuous)  Adverse events – (dichotomous)  Adherence (dichotomous) Importance of  Critical outcomesHbA1c outcomes  Hypoglycaemia Study design RCTs  Unit of randomisation: individual patient Population size  No restrictions on sample size and directness  Exclude studies <1 week duration  Studies with indirect, or mixed diabetes (type 1 diabetes and type 2 diabetes) populations will only be considered, if: o data has been reported for the subgroup of type 1 diabetes patients, in which case this subgroup data will be used, o the population contains ≥70% of type 1 diabetes patients, o if ≥50% of people are aged >18 years the study will be included. Setting  All settings (as per Scope) Search Strategy See appendix F Review Strategy Appraisal of methodological quality  The methodological quality of each study will be assessed using NICE checklists and the quality of the evidence will be assessed by GRADE for each outcome. Synthesis of data  Meta-analysis will be conducted where appropriate  Outcomes will be grouped into the following categories based on time-points: o ≤6 months (or the one nearest to 6 months if multiple time-points are given) o >6 months (or the longest one if multiple time-points are given)  Default MIDs will be used: 0.75 and 1.25 for dichotomous outcomes; 0.5 times SD for continuous outcomes Notes If no/insufficient RCT evidence is found we will (in order of preference):  Consider evidence from non-randomised, non-comparative and observational studies  Consider unpublished or partially published studies (including abstracts – and contact the authors for more information)  Move to GDG consensus

National Clinical Guideline Centre, 2014 55 Type 1 diabetes in adults

Review protocols

C.1.3.8 CGM (continuous) versus CGM (intermittent) Component Description Review question In adults with type 1 diabetes, is continuous real-time monitoring more effective than intermittent real-time monitoring for optimum glucose/diabetic control? Objectives The aim of this review is to determine whether continuous use of real-time CGM is better than intermittent use of real-time CGM for managing diabetic control in adults with type 1 diabetes. Adherence may also be an issue? Population Adults with type 1 diabetes  Adult is defined as aged ≥ 18 years Subgroups The following groups will be considered separately if data are available:  None specified

The following factors will be considered for subgroup analysis if heterogeneity is present:  frequency of CGM  Type of CGM method/monitor (real-time and retrospective) Intervention Intermittent real-time CGM

Comparison Continuous real-time CGM – comparison group must be on similar insulin regimen as the treatment group Outcomes  HbA1c (continuous)  Hypoglycaemia - preferably severe hypoglycaemia if reported (dichotomous or continuous outcome, depending how it is reported)  Quality of life – measured by any measure stated in the paper (continuous)  Adverse events – (dichotomous)  Adherence (dichotomous)

Importance of Critical outcomes outcomes  HbA1c  Hypoglycaemia Study design RCTs, observational (retrospective and prospective cohort studies)  Unit of randomisation: individual patient Population size  No restrictions on sample size and directness  Exclude studies <1 week duration  Studies with indirect, or mixed diabetes (type 1 diabetes and type 2 diabetes) populations will only be considered, if: o data has been reported for the subgroup of type 1 diabetes patients, in which case this subgroup data will be used, o the population contains ≥70% of type 1 diabetes patients, o if ≥50% of people are aged >18 years the study will be included. Setting  All settings (as per Scope) Search Strategy See appendix F Review Strategy Appraisal of methodological quality  The methodological quality of each study will be assessed using NICE checklists  Where possible, the quality of the evidence will be assessed by GRADE for each outcome. Synthesis of data  Meta-analysis will be conducted where appropriate  Outcomes will be grouped into the following categories based on time-points:

National Clinical Guideline Centre, 2014 56 Type 1 diabetes in adults

Review protocols

o ≤6 months (or the one nearest to 6 months if multiple time-points are given) o >6 months (or the longest one if multiple time-points are given)  Default MIDs will be used: 0.75 and 1.25 for dichotomous outcomes; 0.5 times SD for continuous outcomes Notes If no/insufficient RCT evidence is found we will (in order of preference):  Consider evidence from non-randomised, non-comparative and observational studies  Consider unpublished or partially published studies (including abstracts – and contact the authors for more information)  Move to GDG consensus

C.1.4 Insulin therapy

C.1.4.1 Rapid-acting insulin

Review question In adults with type 1 diabetes, which are the most effective rapid-acting insulins for meal times: analogues versus human (intermediate NPH), for optimal diabetic control? Objectives The aim of this review is to determine which is the most effective rapid-acting insulin for use at meal-times for diabetic control in adults with type 1 diabetes. Population Adults with type 1 diabetes  Adult is defined as aged ≥18 years Subgroups The following groups will be considered separately if data are available:  None specified The following factors will be considered for subgroup analysis (in the critical outcomes only) if heterogeneity is present:  different doses of insulin Intervention Rapid-acting insulin: analogues, human (intermediate NPH)  Rapid human  Aspart  Lispro  Glulisine

Note: only UK licensed interventions and doses will be considered Comparison Each other

Note: only UK licensed interventions and doses will be considered Outcomes  HbA1c (continuous)  Hypoglycaemia (dichotomous or continuous outcome, depending how it is reported)  Severe hypoglycaemia (dichotomous or continuous outcome, depending how it is reported)  Nocturnal hypoglycaemia (dichotomous or continuous outcome, depending how it is reported)  Quality of life – measured by DQoL or any measure used in the studies retrieved (continuous)  Patient satisfaction (dichotomous or continuous outcome, depending how it is reported)  Adverse events – Cancer (dichotomous)  Injection site issues (dichotomous or continuous outcome, depending how it is reported)  Weight gain/loss (continuous)

National Clinical Guideline Centre, 2014 57 Type 1 diabetes in adults

Review protocols

 DKA (dichotomous or continuous outcome, depending how it is reported) Importance of Critical outcomes outcomes  HbA1c  Hypoglycaemia Study design RCTs  Unit of randomisation: individual patient Population size  No restrictions on sample size and directness  Studies with a follow-up time of <4 weeks will be excluded.  Studies with indirect, or mixed diabetes (type 1 diabetes and type 2 diabetes) populations will only be considered, if: o data has been reported for the subgroup of type 1 diabetes patients, in which case this subgroup data will be used, o the population contains ≥70% of type 1 diabetes patients, o if ≥50% of people are aged >18 years the study will be included.

Setting  All settings (as per Scope) Search Strategy See appendix F Review Strategy Appraisal of methodological quality  The methodological quality of each study will be assessed using NICE checklists and the quality of the evidence will be assessed by GRADE for each outcome. Synthesis of data  Meta-analysis will be conducted where appropriate  Outcomes will be grouped into the following categories based on time-points: o ≤6 months (or the one nearest to 6 months if multiple time-points are given) o >6 months (or the longest one if multiple time-points are given)  Default MIDs will be used: 0.75 and 1.25 for dichotomous outcomes; 0.5 times SD for continuous outcomes Notes If no/insufficient RCT evidence is found we will (in order of preference):  Consider evidence from non-randomised, non-comparative and observational studies  Consider unpublished or partially published studies (including abstracts – and contact the authors for more information)  Move to GDG consensus

C.1.4.2 Long-acting insulin Component Description Review question In adults with type 1 diabetes, what are the most effective long-acting insulins (detemir versus degludec versus glargine versus NPH) for optimal diabetic control? Objectives The aim of this review is to determine which is the most effective long-acting insulin in adults with type 1 diabetes. Population Adults with type 1 diabetes  Adult is defined as aged ≥18 years Subgroups The following groups will be considered separately if data are available:  None specified upfront

The following factors will be considered for subgroup analysis (in the critical outcomes only) if heterogeneity is present:  Clinically relevant regiment (for example, once or twice/day) Baseline HbA1c (if there are vast differences between studies in baseline HbA1c levels)  Different doses/regimens (clinically relevant regimens)

National Clinical Guideline Centre, 2014 58 Type 1 diabetes in adults

Review protocols

 Elderly/older people/frailty (if there are vast differences between studies in ages)  Baseline weight (may not be possible to do this, because some studies give BMI and some give weight in kg)  Baseline hypoglycaemia (if this is known and there are vast differences) Intervention Long-acting insulins:  detemir  degludec  glargine  NPH/other intermediate

Note: only UK licensed interventions and doses will be considered Comparison Each other (all of the above)

Note: only UK licensed interventions and doses will be considered Outcomes Outcomes  HbA1c (continuous)  Hypoglycaemia - preferably severe hypoglycaemia if reported (dichotomous or continuous outcome, depending how it is reported)  Quality of life – measured by DQoL or any measure used in the studies retrieved (continuous)  Adverse events – Cancer (dichotomous)  Injection site issues  Weight gain/loss  DKA Importance of Critical outcomes outcomes  HbA1c  Hypoglycaemia Study design RCTs  Unit of randomisation: individual patient Population size  No restrictions on sample size (unless have >10 papers for any type of insulin, then and directness sample sizes N<100 will be excluded within that category)  Studies with a follow-up time <4 weeks will be excluded.  Studies with indirect, or mixed diabetes (type 1 diabetes and type 2 diabetes) populations will NOT be considered, unless data has been reported for the subgroup of type 1 diabetes patients, in which case this subgroup data will be used.  Studies only comparing different dosages of the same drug will not be included (for example, NPH 100 mg versus NPH 200 mg) Setting  All settings (as per Scope) Search Strategy See appendix F Review Strategy Appraisal of methodological quality  The methodological quality of each study will be assessed using NICE checklists  Where possible, the quality of the evidence will be assessed by GRADE for each outcome. Synthesis of data  Meta-analysis will be conducted where appropriate  Outcomes will be grouped into the following categories based on time-points: o ≤6 months (or the one nearest to 6 months if multiple time-points are given) o >6 months (or the longest one if multiple time-points are given)  Default MIDs will be used: 0.75 and 1.25 for dichotomous outcomes; 0.5 times SD for

National Clinical Guideline Centre, 2014 59 Type 1 diabetes in adults

Review protocols

continuous outcomes  A network meta-analysis will be conducted for each of the critical outcomes (HbA1cand severe hypoglycaemia). Notes If no/insufficient RCT evidence is found we will (in order of preference):  Consider evidence from non-randomised, non-comparative and observational studies  Consider unpublished or partially published studies (including abstracts – and contact the authors for more information)  Move to GDG consensus

C.1.4.3 Mixed insulin Component Description Review question In adults with type 1 diabetes, what are the most effective mixed insulins (degludec- aspart versus glargine versus NPH) for optimal diabetic control? Objectives The aim of this review is to determine which is the most effective mixed insulin in adults with type 1 diabetes. Population Adults with type 1 diabetes  Adult is defined as aged ≥18 years Subgroups The following groups will be considered separately if data are available:  None specified

The following factors will be considered for subgroup analysis (in the critical outcomes only) if heterogeneity is present:  Baseline HbA1c (if there are vast differences between studies in baseline HbA1clevels)  Different doses or regimens (clinically relevant regimens)  Elderly/older people/frailty (if there are vast differences between studies in ages)  Baseline weight (may not be possible to do this, because some studies give BMI and some give weight in kg)  Baseline hypoglycaemia (if this is known and there are vast differences) Intervention Mixed insulins:  Degludec-aspart – (IDegAsp) NPH-aspart (NovoMix 30)- analogue mix  NPH-lispro (Humalog Mix 25 and Humalog Mix 50)- analogue mixes  NPH-human (Humulin M3, Insuman Combo 25 and Insuman Combo 50)  BD premix – the below repeat the above mixes that are available Comparison  Each other (all of the above)  Long- and short-acting insulin (basal-bolus) regimen Outcomes  HbA1c (continuous)  Hypoglycaemia (dichotomous or continuous outcome, depending how it is reported)  Severe hypoglycaemia (dichotomous or continuous outcome, depending how it is reported)  Nocturnal hypoglycaemia (dichotomous or continuous outcome, depending how it is reported)  Quality of life – measured by DQoL or any measure used in the studies retrieved (continuous)  Adverse events – Cancer (dichotomous)  Injection site issues (dichotomous)  Weight gain/loss (continuous)  DKA (dichotomous or continuous outcome, depending how it is reported) Importance of Critical outcomes

National Clinical Guideline Centre, 2014 60 Type 1 diabetes in adults

Review protocols

outcomes  HbA1c  Hypoglycaemia Study design RCTs  Unit of randomisation: individual patient Population size  No restrictions on sample size (unless have >10 papers for each type of drug, then and directness sample sizes N<100 will be excluded within that category)  Studies with a follow-up time <4 weeks will be excluded.  Studies with indirect, or mixed diabetes (type 1 diabetes and type 2 diabetes) populations will only be considered, if: o data has been reported for the subgroup of type 1 diabetes patients, in which case this subgroup data will be used. o the population contains ≥70% of type 1 diabetes patients o if ≥50% of people are aged >18 years the study will be included  Studies only comparing different dosages of the same drug will not be included (for example, NPH 100 mg versus NPH 200 mg) Setting  All settings (as per Scope) Search Strategy See appendix F Review Strategy Appraisal of methodological quality  The methodological quality of each study will be assessed using NICE checklists  Where possible, the quality of the evidence will be assessed by GRADE for each outcome. Synthesis of data  Meta-analysis will be conducted where appropriate  Outcomes will be grouped into the following categories based on time-points: o ≤6 months (or the one nearest to 6 months if multiple time-points are given) o >6 months (or the longest one if multiple time-points are given)  Default MIDs will be used: 0.75 and 1.25 for dichotomous outcomes; 0.5 times SD for continuous outcomes  A network meta-analysis will be conducted for each of the critical outcomes (HbA1c and severe hypoglycaemia). Notes If no/insufficient RCT evidence is found we will (in order of preference):  Consider evidence from non-randomised, non-comparative and observational studies  Consider unpublished or partially published studies (including abstracts – and contact the authors for more information)  Move to GDG consensus

C.1.4.4 Adjuncts Component Description Review question In adults with type 1 diabetes, is metformin (with or without insulin), or GLP1-agonists (with or without insulin) as effective as insulin alone for optimal diabetic control? Objectives The aim of this review is to determine whether insulin in combination with another pharmacological agent (that is, Metformin or GLP-1 agonists) is as effective as insulin alone for diabetic control in adults with type 1 diabetes. Also to determine the extra benefits such as decreased insulin doses (30  Combination therapy

National Clinical Guideline Centre, 2014 61 Type 1 diabetes in adults

Review protocols

 Monotherapy

The following factors will be considered for subgroup analysis (in the critical outcomes only) if heterogeneity is present:  different doses of insulin, metformin or GLP1-agonist  different frequencies of administration Intervention  Metformin  Metformin + insulin  GLP-1 agonists o exenatide o pramlintide o liraglutide  GLP1 + insulin

Note: only UK licensed interventions and doses will be considered Comparison Insulin

Note: only UK licensed interventions and doses will be considered Outcomes  HbA1c (continuous)  Hypoglycaemia - preferably severe hypoglycaemia if reported (dichotomous or continuous outcome, depending how it is reported)  Severe hypoglycaemia (dichotomous or continuous outcome, depending how it is reported)  Nocturnal hypoglycaemia (dichotomous or continuous outcome, depending how it is reported)  Quality of life – (continuous)  Adverse events – (dichotomous)  Weight loss/change – (dichotomous)  Dose of insulin – (dichotomous) Importance of Critical outcomes outcomes  HbA1c  Hypoglycaemia  Weight loss/change Study design RCTs  Unit of randomisation: individual patient Population size  No restrictions on sample size and directness  Studies with a follow-up time <4 weeks will be excluded.  Studies with indirect, or mixed diabetes (type 1 diabetes and type 2 diabetes) populations will only be considered, if: o data has been reported for the subgroup of type 1 diabetes patients, in which case this subgroup data will be used, o the population contains ≥70% of type 1 diabetes patients, o if ≥50% of people are aged >18 years the study will be included. Setting  All settings (as per Scope) Search Strategy See appendix F Review Strategy Appraisal of methodological quality  The methodological quality of each study will be assessed using NICE checklists and the quality of the evidence will be assessed by GRADE for each outcome. Synthesis of data

National Clinical Guideline Centre, 2014 62 Type 1 diabetes in adults

Review protocols

 Meta-analysis will be conducted where appropriate  Outcomes will be grouped into the following categories based on time-points: o ≤6 months (or the one nearest to 6 months if multiple time-points are given) o >6 months (or the longest one if multiple time-points are given)  Default MIDs will be used: 0.75 and 1.25 for dichotomous outcomes; 0.5 times SD for continuous outcomes Notes If no/insufficient RCT evidence is found we will (in order of preference):  Consider evidence from non-randomised, non-comparative and observational studies  Consider unpublished or partially published studies (including abstracts – and contact the authors for more information)  Move to GDG consensus

C.1.4.5 Long-acting: once versus twice daily basal insulin administration Component Description Review question In adults with type 1 diabetes, is once daily basal insulin more effective than twice daily basal insulin for optimal diabetic control? Objectives The aim of this review is to determine the most effective administration regimen for basal/long-acting insulin: once daily versus twice daily for diabetic control in adults with type 1 diabetes. Population Adults with type 1 diabetes  Adult is defined as aged ≥18 years Subgroups The following groups will be considered separately if data are available:  Age (if available)  Duration of diabetes (<15 years and >15 years)  Control (that is, HbA1c <8% and >8%)

The following factors will be considered for subgroup analysis (in the critical outcomes only) if heterogeneity is present:  different drugs  different doses Intervention Basal/long-acting insulin given once/day: detemir, degludec, degludec/aspart, glargine, NPH

Note: only UK licensed interventions and doses will be considered Comparison Basal/long-acting insulin given twice/day: detemir, degludec, degludec/aspart, glargine, NPH

Note: the same LA insulin must be used in the comparison arm as for the intervention arm. Note: only UK licensed interventions and doses will be considered Outcomes  HbA1c (continuous)  Hypoglycaemia - preferably severe hypoglycaemia if reported (dichotomous or continuous outcome, depending how it is reported)  Quality of life – measured by whatever is used in the study (continuous)  Adverse events – (dichotomous) Importance of Critical outcomes outcomes  HbA1c  Hypoglycaemia Study design RCTs

National Clinical Guideline Centre, 2014 63 Type 1 diabetes in adults

Review protocols

 Unit of randomisation: individual patient Population size  No restrictions on sample size and directness  Studies with a follow-up time <4 weeks will be excluded.  Studies with indirect, or mixed diabetes (type 1 diabetes and type 2 diabetes) populations will NOT be considered, unless data has been reported for the subgroup of type 1 diabetes patients, in which case this subgroup data will be used. Setting  All settings (as per Scope)

Search Strategy See appendix F Review Strategy Appraisal of methodological quality  The methodological quality of each study will be assessed using NICE checklists  Where possible the quality of the evidence will be assessed by GRADE for each outcome. Synthesis of data  Meta-analysis will be conducted where appropriate  Outcomes will be grouped into the following categories based on time-points: o ≤6 months (or the one nearest to 6 months if multiple time-points are given) o >6 months (or the longest one if multiple time-points are given)  Default MIDs will be used to assess imprecision: 0.75 and 1.25 for dichotomous outcomes; 0.5 times SD for continuous outcomes Notes If no/insufficient RCT evidence is found we will (in order of preference):  Consider evidence from non-randomised, non-comparative and observational studies  Consider unpublished or partially published studies (including abstracts – and contact the authors for more information)  Move to GDG consensus

C.1.5 Insulin delivery

C.1.5.1 Needle length Component Description Review question In adults with type 1 diabetes, what is the optimum needle length for insulin delivery? Objectives The aim of this review is to determine the most effective needle length for administering insulin in adults with type 1 diabetes. Issues of adherence? Population Adults with type 1 diabetes  Adult is defined as aged ≥ 18 years Subgroups The following groups will be considered separately if data are available:  None specified

The following factors will be considered for subgroup analysis (in the critical outcomes only) if heterogeneity is present:  different types of needle  different frequencies of insulin delivery  different sites of delivery  different insulin given Intervention Insulin – delivered by needle

Note: only UK licensed interventions and doses will be considered Comparison As for intervention, but different length of needle

National Clinical Guideline Centre, 2014 64 Type 1 diabetes in adults

Review protocols

Note: only UK licensed interventions and doses will be considered Outcomes Outcomes  Pain (continuous)  Discomfort (continuous)  Patient satisfaction (continuous)  HbA1c (continuous)  Quality of life – measured by sale reported in the study (continuous)  Adverse events (dichotomous)  Adherence (dichotomous) Importance of Critical outcomes outcomes  HbA1c Study design RCTs, observational (retrospective and prospective cohort studies)  Unit of randomisation: individual patient  Studies will be excluded if they are non-comparative studies Population size  No restrictions on sample size and directness  No restrictions on treatment duration  Studies with indirect, or mixed diabetes (type 1 diabetes and type 2 diabetes) populations will only be considered, if: o data has been reported for the subgroup of type 1 diabetes patients, in which case this subgroup data will be used. o the population contains ≥70% of type 1 diabetes patients o if ≥50% of people are aged >18 years the study will be included  However, if very little evidence is found meeting the above population criteria, studies with mixed diabetes (type 1 diabetes and type 2 diabetes) populations will be considered, regardless of the % of type 1 diabetes patients.  Studies will be excluded if they compare insulin pens with vial and syringe. Delivery of insulin must be by the same type of insulin ‘device’ in order to show the effect of the needle length only, rather than the device. Setting  All settings (as per Scope) Search Strategy See appendix F Review Strategy Appraisal of methodological quality  The methodological quality of each study will be assessed using NICE checklists  Where possible, the quality of the evidence will be assessed by GRADE for each outcome. Synthesis of data  Meta-analysis will be conducted where appropriate  Outcomes will be grouped into the following categories based on time-points: o ≤6 months (or the one nearest to 6 months if multiple time-points are given) o >6 months (or the longest one if multiple time-points are given)  Default MIDs will be used: 0.75 and 1.25 for dichotomous outcomes; 0.5 times SD for continuous outcomes Notes If no/insufficient RCT evidence is found we will (in order of preference):  Consider evidence from non-randomised, non-comparative and observational studies  Consider unpublished or partially published studies (including abstracts – and contact the authors for more information)  Move to GDG consensus

C.1.5.2 Site and rotation Component Description

National Clinical Guideline Centre, 2014 65 Type 1 diabetes in adults

Review protocols

Review question In adults with type 1 diabetes, what is the optimum injection site and rotation for insulin delivery? Objectives The aim of this review is to determine which is the most effective delivery site and rotation for administering insulin in adults with type 1 diabetes. Issues of adherence? Population Adults with type 1 diabetes  Adult is defined as aged ≥ 18 years Subgroups The following groups will be considered separately if data are available:  Pregnancy  Exercise  Nocturnal hypoglycaemia

The following factors will be considered for subgroup analysis (in the critical outcomes only) if heterogeneity is present:  different types/lengths of needle  different frequencies of insulin delivery  different doses Intervention Insulin – delivered by needle

Note: only UK licensed interventions and doses will be considered Comparison As for intervention, but different site of delivery

Note: only UK licensed interventions and doses will be considered Outcomes Outcomes  HbA1c (continuous)  Hypoglycaemia (dichotomous or continuous outcome, depending how it is reported)  Severe hypoglycaemia (dichotomous or continuous outcome, depending how it is reported)  Nocturnal hypoglycaemia (dichotomous or continuous outcome, depending how it is reported)  Quality of life – measured by whatever is used in the study (continuous)  Adverse events – (dichotomous)  Adherence (dichotomous) Importance of Critical outcomes outcomes  HbA1c  Hypoglycaemia Study design RCTs, observational (retrospective and prospective cohort studies)  Unit of randomisation: individual patient Population size  No restrictions on sample size and directness  No restrictions on treatment duration  Studies with indirect, or mixed diabetes (type 1 diabetes and type 2 diabetes) populations will only be considered, if: o data has been reported for the subgroup of type 1 diabetes patients, in which case this subgroup data will be used. o the population contains ≥70% of type 1 diabetes patients o if ≥50% of people are aged >18 years the study will be included Setting  All settings (as per Scope) Search Strategy See appendix F Review Strategy Appraisal of methodological quality

National Clinical Guideline Centre, 2014 66 Type 1 diabetes in adults

Review protocols

 The methodological quality of each study will be assessed using NICE checklists  Where possible, the quality of the evidence will be assessed by GRADE for each outcome. Synthesis of data  Meta-analysis will be conducted where appropriate  Outcomes will be grouped into the following categories based on time-points: o ≤6 months (or the one nearest to 6 months if multiple time-points are given) o >6 months (or the longest one if multiple time-points are given)  Default MIDs will be used: 0.75 and 1.25 for dichotomous outcomes; 0.5 times SD for continuous outcomes Notes If no/insufficient RCT evidence is found we will (in order of preference):  Consider evidence from non-randomised, non-comparative and observational studies  Consider unpublished or partially published studies (including abstracts – and contact the authors for more information)  Move to GDG consensus

C.1.6 Pancreas transplant and islet cell transplantation Component Description Review question Which adults with type 1 diabetes are most suitable to be considered for a pancreas transplant, or pancreatic islet cell transplantation? Objectives The aim of this review is to determine the referral criteria indicating that an adult with type 1 diabetes should be considered for pancreas transplant or pancreatic islet cell transplantation. Population Adults with type 1 diabetes  Adult is defined as aged ≥18 years  Type 1 diabetes is defined (WHO definition and NICE 2004 GL) Subgroups The following groups will be considered separately if data are available:  Islet cell transplantation - alone and with kidney transplant  Pancreas transplantation - alone (not in combination with kidney) Intervention  Pancreas transplantation  Islet cell transplantation Comparison  Any comparison  No comparison Outcomes  The referral criteria themselves  HbA1c  Severe hypoglycaemia  Longevity of the transplant/organ survival (c-peptide and insulin independence)  Insulin dependence at 1 year and 5 years  Mortality - in-hospital/procedural  Mortality – long-term  Quality of life – any measure used in the paper Importance of Critical outcomes outcomes  Mortality  Severe hypoglycaemia  Longevity of transplant  Insulin dependence/independence

National Clinical Guideline Centre, 2014 67 Type 1 diabetes in adults

Review protocols

Study Strategy for this review is: design/review  Use available clinical data for obtaining clinical outcomes. This will be sourced from: strategy o Publications of the UK consortia data (National transplant programmes); to see whether patients do well after a transplant which is based on the current UK referral criteria. o NICE IPG 257 guidance on transplantation.  Report what referral criteria are currently used in the UK. This will be sourced from: o NHS England service specifications document o NICE IPG 257 guidance for any information given on referral criteria.  International criteria : CITR (data can be found on their website and any relevant publications) Population size  Information on transplantation before 2003 will be excluded (as not current/relevant and directness technology for the procedures).  Studies with outcomes at <5 years will be excluded unless there is not much other data.  No restrictions on sample size  No restrictions on treatment duration  Studies with indirect, or mixed diabetes (type 1 diabetes and type 2 diabetes) populations will only be considered, if: o data has been reported for the subgroup of type 1 diabetes patients, in which case this subgroup data will be used. o the population contains ≥70% of type 1 diabetes patients o if ≥50% of people are aged >18 years the study will be included Setting  All settings (as per Scope) Search Strategy See study design/review strategy section above Review Strategy Synthesis of data  Narrative summary of the data will be given, as there will be no RCT or trial data for this review. Notes If no/insufficient evidence is found we will move to GDG consensus

C.1.7 Hypoglycaemia

C.1.7.1 Identification and quantification of impaired awareness of hypoglycaemia Component Description Review question In adults with type 1 diabetes, how is impaired awareness of hypoglycaemia best identified and quantified? Objectives The aim of this review is to look at how effective are scoring systems in predicting impaired awareness of hypoglycaemia and increased risk of severe hypoglycaemia. Also how frequently should we be monitoring for it, and using which tool. What is the quickest and most reliable tool? Population  Adults with type 1 diabetes o Adult is defined as aged ≥18 years Subgroups None identified. Prognostic  Impaired awareness of hypoglycaemia according to known validated scoring systems: variable? o Gold score o Clarke score o Ryan score (Hypoglycaemia burden score)

National Clinical Guideline Centre, 2014 68 Type 1 diabetes in adults

Review protocols

o Pedersen-Bjergaard score Comparison  Other scoring systems  No scoring system Outcomes  Ability to identify impaired awareness of hypoglycaemia  Ability to predict severe hypoglycaemia (incidence of severe hypoglycaemia)  Ability to predict driving or work related accidents (incidence of accidents) Importance of Critical outcomes outcomes  Ability to predict severe hypoglycaemia Study design All study types Population size  No restrictions on sample size and directness  No restrictions on study duration  Studies with indirect, or mixed diabetes (type 1 diabetes and type 2 diabetes) populations will only be considered, if: o data have been reported for the subgroup of type 1 diabetes patients, in which case these subgroup data will be used. o the population contains ≥70% of type 1 diabetes patients o if ≥50% of people are aged >18 years the study will be included Setting  All settings (as per Scope) Search Strategy  See appendix F Review Strategy  Appraisal of methodological quality  The methodological quality of each study will be assessed using NICE checklists and the quality of the evidence will be assessed by GRADE for each outcome. Notes  If no/insufficient RCT evidence is found we will (in order of preference): o Consider evidence from non-randomised, non-comparative and observational studies o Consider unpublished or partially published studies (including abstracts – and contact the authors for more information) o Move to GDG consensus

C.1.7.2 Recovering hypoglycaemia awareness In adults with type 1 diabetes and impaired awareness of hypoglycaemia, what is the Review question most effective strategy for recovering hypoglycaemia awareness? Objectives The aim of this review is to look at what are the most effective methods to increase awareness of hypoglycaemia in people with type 1 diabetes and impaired awareness of hypoglycaemia. Population Adults with type 1 diabetes and with impaired awareness of hypoglycaemia  Adult is defined as aged ≥18 years Subgroups The following factors will be considered for subgroup analysis (in the critical outcomes only) if heterogeneity is present, in the following order:  Age (>60 years)  Disease duration (>15 years)  Differing insulin regimens – those on MDI, bd. or PUMP  Duration of unawareness (>6 months)  People with previous hypoglycaemia unawareness versus those without  HbA1c (>7.5%)

National Clinical Guideline Centre, 2014 69 Type 1 diabetes in adults

Review protocols

Intervention  Adjusting treatment/adjusting insulin regime/less intensive glycaemic control  Pancreas/islet cell transplant  Hypoglycaemia avoidance  Adjusting monitoring of blood glucose for example CGM  Education interventions  Psychological interventions  Treatments that bring back the warning signs of hypoglycaemia? (no specific treatments have been suggested yet)

Only intervention durations of ≥1 month will be considered Comparison  Any  None Outcomes  HbA1c (continuous)  Autonomic symptoms/symptom scores during hypoglycaemia clamp study  Hypoglycaemia (dichotomous or continuous outcome, depending how it is reported)  Severe hypoglycaemia (dichotomous or continuous outcome, depending how it is reported)  Nocturnal hypoglycaemia (dichotomous or continuous outcome, depending how it is reported)  Hospital admissions (dichotomous or continuous, depending how it is reported)  Hypoglycaemia unawareness or awareness (dichotomous or continuous, depending how it is reported)  Quality of life – measured by DQoL, DSQoL, PAID, HADS, fear of hypoglycaemia, anxiety, depression, cognitive function (continuous)  Road traffic accidents and work related accidents Importance of Critical outcomes outcomes  HbA1c  Severe hypoglycaemia  Hypoglycaemia unawareness or awareness  Quality of life Study design  RCTs: unit of randomisation: individual patient, cluster randomised trials  Observational studies Population size  No restrictions on sample size and directness  Studies with a follow-up time <4 weeks will be excluded.  Studies with indirect, or mixed diabetes (type 1 diabetes and type 2 diabetes) populations will only be considered, if: o data have been reported for the subgroup of type 1 diabetes patients, in which case this subgroup data will be used. o the population contains ≥70% of type 1 diabetes patients o if ≥50% of people are aged >18 years the study will be included Setting  All settings (as per Scope) Search Strategy See appendix F Review Strategy Appraisal of methodological quality  The methodological quality of each study will be assessed using NICE checklists and

National Clinical Guideline Centre, 2014 70 Type 1 diabetes in adults

Review protocols

the quality of the evidence will be assessed by GRADE for each outcome. Synthesis of data  Meta-analysis will be conducted where appropriate  Outcomes will be grouped into the following categories based on time-points: o ≤6 months (or the one nearest to 6 months if multiple time-points are given) o >6 months (or the longest one if multiple time-points are given)  Default MIDs will be used: 0.75 and 1.25 for dichotomous outcomes; 0.5 times SD for continuous outcomes Notes If no/insufficient RCT evidence is found we will (in order of preference):  Consider evidence from non-randomised, non-comparative and observational studies  Consider unpublished or partially published studies (including abstracts – and contact the authors for more information)  Move to GDG consensus

C.1.8 Ketone monitoring

C.1.8.1 Ketone self-monitoring Component Description Review questions In adults with type 1 diabetes (including atypical ketosis-prone diabetes), does patient self-monitoring of blood (and urine) ketones reduce the incidence of DKA and hospital admissions? Objectives The aim of this review is to determine whether ketone monitoring is an effective method to help prevent DKA occurring in adults with type 1 diabetes. Population Adults with type 1 diabetes  Adult is defined as aged ≥ 18 years Subgroups The following groups will be considered separately if data are available:  Atypical ketosis-prone diabetes

The following factors will be considered for subgroup analysis if heterogeneity is present:  Frequency of monitoring  Frequency of DKA  Whether patients had formal education in insulin monitoring  Recurrent DKA or Intermittent DKA  Whether DKA had a clear precipitating factor (for example, heart attack) Intervention and  Blood ketone versus urine ketone measurements comparisons  Any or no comparison Outcomes  Hospital admissions – for DKA if specified (dichotomous)  Duration of admission/length of hospital stay (continuous)  DKA (dichotomous)  HbA1c (continuous)  Hypoglycaemia - preferably severe hypoglycaemia if reported (dichotomous or continuous outcome, depending how it is reported)  Quality of life – measured by PAID, anxiety (continuous)  Severity of acidosis at admission - duration of acidosis and degree of acidosis (continuous or dichotomous if split into categories) Importance of Critical outcomes outcomes

National Clinical Guideline Centre, 2014 71 Type 1 diabetes in adults

Review protocols

 Hospital admissions Study design All study types Population size  No restrictions on sample size and directness  No restrictions on duration  Studies with indirect, or mixed diabetes (type 1 diabetes and type 2 diabetes) populations will not be considered unless ≥75% are type 1 diabetes, or data has been reported for the subgroup of type 1 diabetes patients, in which case this subgroup data will be used.  However, if there are very few studies found (n<5) then we will consider the following indirect populations: o mixed type 1 diabetes and type 2 diabetes populations, (as DKA usually occurs most frequently in type 1 diabetes patients so studies are most likely to have a higher % of type 1 diabetes) o mixed aged diabetes populations (adults, young people and children) Setting  All settings (as per Scope) Search Strategy See appendix F Review Strategy Appraisal of methodological quality  The methodological quality of each study will be assessed using NICE checklists  Where possible, the quality of the evidence will be assessed by GRADE for each outcome. Synthesis of data  Meta-analysis will be conducted where appropriate  Outcomes will be grouped into the following categories based on time-points: o ≤6 months (or the one nearest to 6 months if multiple time-points are given) o >6 months (or the longest one if multiple time-points are given)  Default MIDs will be used: 0.75 and 1.25 for dichotomous outcomes; 0.5 times SD for continuous outcomes Notes If no/insufficient RCT evidence is found we will (in order of preference):  Consider evidence from non-randomised, non-comparative and observational studies  Consider indirect populations (as stated above)  Consider unpublished or partially published studies (including abstracts – and contact the authors for more information)  Move to GDG consensus

C.1.8.2 Ketone monitoring in-hospital Component Description Review questions In adults with type 1 diabetes does in-patient monitoring of blood ketones by the healthcare professional reduce the length of hospital stay, exposure to IV insulin and the development of in-hospital complications:  in patients with suspected DKA?  in patients admitted with DKA and/or those that get it in hospital. Objectives The aim of this review is to determine whether ketone monitoring is an effective method to help prevent DKA occurring in adults with type 1 diabetes, and in those already with DKA does it reduce the length of hospital stay and the development of other adverse events. Population Adults with type 1 diabetes  Adult is defined as aged ≥18 years Subgroups The following groups will be considered separately if data are available:  None specified

National Clinical Guideline Centre, 2014 72 Type 1 diabetes in adults

Review protocols

The following factors will be considered for subgroup analysis if heterogeneity is present:  Frequency of monitoring  Hospital admissions  Ketone measurement: finger strip or lab measurement  Severity of type 1 diabetes and DKA (mild, moderate or severe) Intervention  Blood Ketone monitoring

Note: only UK licensed interventions and doses will be considered Comparison  Urine ketone  No monitoring

Note: only UK licensed interventions and doses will be considered Outcomes Outcomes  Length of hospital stay (continuous)  In-hospital complications of the admission – for example cerebral oedema, mortality, serious electrolyte imbalance (dichotomous)  Exposure to IV insulin (dichotomous)  How often admission occurs (continuous)  HbA1c (continuous)  Hypoglycaemia - preferably severe hypoglycaemia if reported (dichotomous or continuous outcome, depending how it is reported)  Quality of life - (continuous) Importance of Critical outcomes outcomes  Mortality Study design All study types Population size  No restrictions on sample size and directness  No restrictions on duration  Studies with indirect, or mixed diabetes (type 1 diabetes and type 2 diabetes) populations will not be considered unless ≥75% are type 1 diabetes, or data has been reported for the subgroup of type 1 diabetes patients, in which case this subgroup data will be used.  However, if there are very few studies found (n<5) then we will consider indirect populations: o mixed type 1 diabetes and type 2 diabetes populations, (as DKA usually occurs most frequently in type 1 diabetes patients so studies are most likely to have a higher % of type 1 diabetes) o mixed aged diabetes populations (adults, young people and children) Setting  All settings (as per Scope) Search Strategy See appendix F Review Strategy Appraisal of methodological quality  The methodological quality of each study will be assessed using NICE checklists  Where possible, the quality of the evidence will be assessed by GRADE for each outcome. Synthesis of data  Meta-analysis will be conducted where appropriate  Outcomes will be grouped into the following categories based on time-points: o ≤6 months (or the one nearest to 6 months if multiple time-points are given) o >6 months (or the longest one if multiple time-points are given)

National Clinical Guideline Centre, 2014 73 Type 1 diabetes in adults

Review protocols

 Default MIDs will be used: 0.75 and 1.25 for dichotomous outcomes; 0.5 times SD for continuous outcomes Notes If no/insufficient RCT evidence is found we will (in order of preference):  Consider evidence from non-randomised, non-comparative and observational studies  Consider indirect populations (as stated above)  Consider unpublished or partially published studies (including abstracts – and contact the authors for more information)  Move to GDG consensus

C.1.9 Arterial risk control

C.1.9.1 Aspirin Component Description Review question In adults with type 1 diabetes, is aspirin an effective anti-platelet agent for the primary prevention of CV events? Objectives The aim of this review is to determine whether aspirin is an effective agent for preventing CV events in terms of clinical and cost-effectiveness as well as safety for use in adults with type 1 diabetes. Population Adults with type 1 diabetes  Adult is defined as aged ≥ 18 years Subgroups The following groups will be considered separately if data are available:  none specified

The following factors will be considered for subgroup analysis if heterogeneity is present:  aspirin dose  hypertension  micro albuminuria  statin use  PPI use (for adverse event outcomes)  smoking (versus non-smoking) Intervention Aspirin

Note: only UK licensed interventions and doses will be considered Comparison  Placebo  Usual care/no treatment  Low dose versus high dose

Note: only UK licensed interventions and doses will be considered Outcomes Outcomes  Mortality – all cause (dichotomous/time-to event)  Mortality – CV (dichotomous/time-to event)  MI – all cause (dichotomous/time-to event)  MI – fatal (dichotomous/time-to event)  MI – non-fatal (dichotomous/time-to event)  Stroke – all cause (dichotomous/time-to event)  Stroke – fatal (dichotomous/time-to event)  Stroke – non-fatal (dichotomous/time-to event)

National Clinical Guideline Centre, 2014 74 Type 1 diabetes in adults

Review protocols

 Quality of life – measured by SF-36, DQoL, DSQoL (continuous)  Adverse events – bleeding or GI complications (dichotomous)  HbA1c (continuous)  Hypoglycaemia - preferably severe hypoglycaemia if reported (dichotomous or continuous outcome, depending how it is reported)

Importance of Critical outcomes outcomes  Mortality  MI  Stroke Study design RCTs, observational studies  Unit of randomisation: individual patient Population size  No restrictions on sample size and directness  Studies with a follow-up time <4 weeks will be excluded.  Studies with indirect, or mixed diabetes (type 1 diabetes and type 2 diabetes) populations will not be considered, unless data has been reported for the subgroup of type 1 diabetes patients, in which case this subgroup data will be used.  Studies only comparing different dosages of aspirin will not be included Setting  All settings (as per Scope) Search Strategy See appendix F Review Strategy Appraisal of methodological quality  The methodological quality of each study will be assessed using NICE checklists and the quality of the evidence will be assessed by GRADE for each outcome. Synthesis of data  Meta-analysis will be conducted where appropriate  Outcomes will be grouped into the following categories based on time-points: o ≤6 months (or the one nearest to 6 months if multiple time-points are given) o >6 months (or the longest one if multiple time-points are given)  Default MIDs will be used: 0.75 and 1.25 for dichotomous outcomes; 0.5 times SD for continuous outcomes Notes If no/insufficient RCT evidence is found we will (in order of preference):  Consider evidence from non-randomised, non-comparative and observational studies  Consider unpublished or partially published studies (including abstracts – and contact the authors for more information)  Move to GDG consensus

C.1.10 Inpatient management

C.1.10.1 IV insulin (devices and regimens) In adults with type 1 diabetes who have been admitted to hospital (elective and emergency), what is the most effective intravenous insulin dose-adjustment devices Review question and regimens for optimal diabetic control? Objectives The aim of this review is to determine whether IV insulin should be used for treating inpatients with type 1 diabetes, and if so what is the best regimen. Population Adults with type 1 diabetes  Adult is defined as aged ≥18 years Subgroups The following groups will be considered separately if data are available:  Emergency (non-diabetic)  Elective (when nil by mouth)

National Clinical Guideline Centre, 2014 75 Type 1 diabetes in adults

Review protocols

 Surgery  indication groups: DKA or other acute illness, surgical patients, enteral feeding

The following factors will be considered for subgroup analysis (in the critical outcomes only) if heterogeneity is present:  Dose Intervention  IV insulin

Note: only UK licensed interventions and doses will be considered Comparison  Subcutaneous insulin  Each other (different regimens)  Each other (different devices)  No comparison Note: only UK licensed interventions and doses will be considered Outcomes  Achieving target BG levels (dichotomous or continuous outcome, depending how it is reported)  Hypoglycaemia - preferably severe hypoglycaemia if reported (dichotomous or continuous outcome, depending how it is reported)  Time spent out of target glucose, that is, hypoglycaemia/hyperglycaemia (dichotomous or continuous outcome, depending how it is reported)  Duration of IV treatment (continuous)  In-patient stay (continuous)  In-patient mortality (dichotomous)  Infection rate /wound healing (dichotomous or continuous outcome, depending how it is reported)  Quality of life – measured by SF-36, DQoL, DSQoL (continuous) Importance of Critical outcomes outcomes  Achieving target BG levels  Mortality  Hypoglycaemia Study design RCTs, observational studies  Unit of randomisation: individual patient Population size  Exclude studies in ACS populations and directness  No restrictions on sample size  No restrictions on study duration  Studies with indirect, or mixed diabetes (type 1 diabetes and type 2 diabetes) populations will only be considered, if: o data has been reported for the subgroup of type 1 diabetes patients, in which case this subgroup data will be used. o the population contains ≥70% of type 1 diabetes patients o if ≥50% of people are aged >18 years the study will be included Setting  All settings (as per Scope) Search Strategy See appendix F Review Strategy Appraisal of methodological quality  The methodological quality of each study will be assessed using NICE checklists and the quality of the evidence will be assessed by GRADE for each outcome. Synthesis of data  Meta-analysis will be conducted where appropriate  Outcomes will be grouped into the following categories based on time-points:

National Clinical Guideline Centre, 2014 76 Type 1 diabetes in adults

Review protocols

o ≤6 months (or the one nearest to 6 months if multiple time-points are given) o >6 months (or the longest one if multiple time-points are given)  Default MIDs will be used: 0.75 and 1.25 for dichotomous outcomes; 0.5 times SD for continuous outcomes Notes If no/insufficient RCT evidence is found we will (in order of preference):  Consider evidence from non-randomised, non-comparative and observational studies  Consider unpublished or partially published studies (including abstracts – and contact the authors for more information)  Move to GDG consensus

C.1.11 Complications

C.1.11.1 Gastroparesis Component Description Review question In adults with type 1 diabetes, what is the most effective treatment for Gastroparesis? Objectives The aim of this review is to determine which is the most effective treatment for Gastroparesis? Population Adults with type 1 diabetes  Adult is defined as aged ≥18 years Subgroups The following groups will be considered separately if data are available:  young poorly controlled females

The following factors will be considered for subgroup analysis (in the critical outcomes only) if heterogeneity is present:  different doses  different interventions and comparisons  use of PPI Intervention  Prokinetic agents/gastroprokinetic agents (for example, erythromycin)  5-Hydoxytryptamine antagonists (for example, ondansetron)  Anti-emetics  Botulinum toxin  Electrical stimulation interventions  Intensive insulin treatment/glucose control  Dietary changes  Enteral feeding  Acupuncture  Aldose reductase inhibitors (including epalrestat)  Histamine-2 receptor antagonists  Centrally acting antidepressants  Surgical interventions (including gastrectomy)

Note: only UK licensed interventions and doses will be considered Comparison  Placebo  Standard care  Each other (within class and between-class comparisons)  Continuous agent versus other agents  Rotation of medications

National Clinical Guideline Centre, 2014 77 Type 1 diabetes in adults

Review protocols

Note: only UK licensed interventions and doses will be considered Outcomes  Hospital admissions (dichotomous)  Severe hypoglycaemia (dichotomous or continuous outcome, depending how it is reported)  Vomiting (dichotomous or continuous outcome, depending how it is reported)  Weight loss (continuous)  Quality of Life - SF-36 (continuous)  HbA1c (continuous)  Symptom control (as defined by the study; dichotomous or continuous outcome, depending how it is reported) Importance of Critical outcomes outcomes  Hospital admissions  Severe hypoglycaemia  Vomiting Study design RCTs Unit of randomisation: individual patient Population size  No restrictions on sample size and directness  Studies with a treatment duration of ≤1 day will be excluded  Studies with indirect, or mixed diabetes (type 1 diabetes and type 2 diabetes) populations will be considered if: o ≥50% of people are aged >18 years the study will be included o studies will be included if any percentage of type 1 diabetes is in a mixed diabetes population (because gastroparesis treatment is not dependent upon/affected by the type of diabetes).  Studies only comparing different dosages or regimens of the same intervention will be excluded  Studies looking at cisapride will be excluded as this is no longer used in the UK.  Studies looking at constipation or gastric emptying (in diabetics who do not have gastroparesis) will be excluded Setting  All settings (as per Scope) Search Strategy See appendix F Review Strategy  To find data published in 2003 and before, we will look for studies in the old guideline and search reference lists of systematic reviews which have included these dates. - Appraisal of methodological quality  The methodological quality of each study will be assessed using NICE checklists and the quality of the evidence will be assessed by GRADE for each outcome. Synthesis of data  Meta-analysis will be conducted where appropriate  Outcomes will be grouped into the following categories based on time-points: o ≤6 months (or the one nearest to 6 months if multiple time-points are given) o >6 months (or the longest one if multiple time-points are given)  Default MIDs will be used: 0.75 and 1.25 for dichotomous outcomes; 0.5 times SD for continuous outcomes Notes If no/insufficient RCT evidence is found we will (in order of preference):  Consider evidence from non-randomised, non-comparative and observational studies (prospective studies only)  Consider unpublished or partially published studies (including abstracts – and contact the authors for more information)  Move to GDG consensus

National Clinical Guideline Centre, 2014 78 Type 1 diabetes in adults

Review protocols

C.1.11.2 Acute painful neuropathy In adults with type 1 diabetes, what is the most effective treatment for acute painful Review question neuropathy? Objectives The aim of this review is to determine which is the most effective treatment for acute painful neuropathy. Population Adults with type 1 diabetes  Adult is defined as aged ≥18 years Subgroups The following groups will be considered separately if data are available:  Newly diagnosed (up to 3 months)

The following factors will be considered for subgroup analysis (in the critical outcomes only) if heterogeneity is present:  Different doses Intervention  Analgesia, for example, Duloxetine, tramadol  Anti-epileptics, a-depressants - tricyclic antidepressants (SNRIs and duloxetine), anti- convulsants (gabapentin, pregabalin), pump therapy  Lidocaine/lignocaine (anaesthetics); capsaicin.  Using pump

Note: only UK licensed interventions and doses will be considered Comparison  Anything  None

Note: only UK licensed interventions and doses will be considered Outcomes  Pain scores (continuous)  Retinopathy – incidence (dichotomous)  Low-level (micro) albuminuria - incidence (dichotomous)  Resolution of symptoms (continuous)  Improvement in pain scores (dichotomous) Importance of Critical outcomes outcomes  Pain scores  Time to resolution of symptoms (continuous) Study design All study types Population size  No restrictions on sample size and directness  No restrictions on treatment duration  Studies with indirect, or mixed diabetes (type 1 diabetes and type 2 diabetes) populations will only be considered, if: o data has been reported for the subgroup of type 1 diabetes patients, in which case this subgroup data will be used. o the population contains ≥70% of type 1 diabetes patients o if ≥50% of people are aged >18 years the study will be included Setting  All settings (as per Scope) Search Strategy See Appendix F Review Strategy Appraisal of methodological quality The methodological quality of each study will be assessed using NICE checklists and the quality of the evidence will be assessed by GRADE for each outcome. Synthesis of data  Meta-analysis will be conducted where appropriate

National Clinical Guideline Centre, 2014 79 Type 1 diabetes in adults

Review protocols

In adults with type 1 diabetes, what is the most effective treatment for acute painful Review question neuropathy?  Outcomes will be grouped into the following categories based on time-points: o ≤6 months (or the one nearest to 6 months if multiple time-points are given) o >6 months (or the longest one if multiple time-points are given)  Default MIDs will be used: 0.75 and 1.25 for dichotomous outcomes; 0.5 times SD for continuous outcomes Notes If insufficient RCT evidence is found we will (in order of preference):  Consider evidence from non-randomised, non-comparative and observational studies  Consider unpublished or partially published studies (including abstracts – and contact the authors for more information)  Move to GDG consensus

C.1.11.3 Thyroid disease monitoring Component Description Review question How should adults with type 1 diabetes be monitored for thyroid disease, and how frequently Objectives The aim of this review is to determine whether and how often thyroid disease should be monitored for, as this is an association of type 1 diabetes Population Adults with type 1 diabetes  Adult is defined as aged ≥18 years Subgroups The following groups will be considered separately if data are available:  different monitoring strategies  different Abs – positive versus not positive

The following factors will be considered for subgroup analysis (in the critical outcomes only) if heterogeneity is present:  Annual versus 5 yearly for hypo- and hyperthyroidism (positive peroxidase Abs)  Did the test (in the paper) specify which method was used  gender (male/female)  family history (yes/no) Intervention Thyroid disease monitoring:  Thyroid function tests  Autoantibodies/antibodies (for example, peroxidase) Comparison  As for the intervention but at a different frequency  Standard care/no monitoring  No comparison (non-comparative studies) Outcomes  Detection of thyroid disease – thyroid tests, for example, TSH, T4 (mainly prevalence)  Incidence of thyroid disease (mainly prevalence)  Frequency of treatment Importance of Critical outcomes outcomes  All of the above Study design  RCTs, observational studies, prognostic studies Population size  No restrictions on sample size (unless >10 studies, then sample sizes of N<100 will be and directness excluded)  No restriction on study duration  Not pregnancy

National Clinical Guideline Centre, 2014 80 Type 1 diabetes in adults

Review protocols

 Studies with indirect, or mixed diabetes (type 1 diabetes and type 2 diabetes) populations will only be considered, if: o data has been reported for the subgroup of type 1 diabetes patients, in which case this subgroup data will be used. o the population contains ≥70% of type 1 diabetes patients o if ≥50% of people are aged >18 years the study will be included Setting  All settings (as per Scope) Search Strategy See appendix F Review Strategy Appraisal of methodological quality  The methodological quality of each study will be assessed using NICE checklists  Where possible, the quality of the evidence will be assessed by GRADE for each outcome.  Taking into consideration the advice on prognostic reviews in the NICE guidelines manual, meta-analysis or GRADE will not be undertaken for prognostic studies Synthesis of data  Meta-analysis will be conducted where appropriate  Data will be synthesised in a narrative review where meta-analysis is not possible  Outcomes will be grouped into the following categories based on time-points: o ≤6 months (or the one nearest to 6 months if multiple time-points are given) o >6 months (or the longest one if multiple time-points are given)  Default MIDs will be used: 0.75 and 1.25 for dichotomous outcomes; 0.5 times SD for continuous outcomes Notes If no/insufficient RCT, observational or prognostic evidence is found, we will (in order of preference):  Consider unpublished or partially published studies (including abstracts – and contact the authors for more information)  Move to GDG consensus

C.2 Health economic review Review All questions – health economic evidence question Objectives To identify economic evaluations relevant to any of the review questions. Search  Populations, interventions and comparators must be as specified in the individual review criteria protocol above.  Studies must be of a relevant economic study design (cost–utility analysis, cost-effectiveness analysis, cost–benefit analysis, cost–consequences analysis, comparative cost analysis).  Studies must not be an abstract only, a letter, editorial or commentary, or a review of economic evaluations. (Recent reviews will be ordered although not reviewed. The bibliographies will be checked for relevant studies, which will then be ordered.)  Unpublished reports will not be considered unless submitted as part of a call for evidence.  Studies must be in English.  Studies must not be published before 1999. Search An economic study search will be undertaken using population-specific terms and an economic strategy study filter – see Appendix F. Review Each study fulfilling the criteria above will be assessed for applicability and methodological strategy limitations using the NICE economic evaluation checklist which can be found in Appendix G of the NICE guidelines manual (2012).1

Inclusion and exclusion criteria

National Clinical Guideline Centre, 2014 81 Type 1 diabetes in adults

Review protocols

 If a study is rated as both ‘Directly applicable’ and with ‘Minor limitations’ then it will be included in the guideline. An economic evidence table will be completed and it will be included in the economic evidence profile.  If a study is rated as either ‘Not applicable’ or with ‘Very serious limitations’ then it will usually be excluded from the guideline. If it is excluded then an economic evidence table will not be completed and it will not be included in the economic evidence profile.  If a study is rated as ‘Partially applicable’, with ‘Potentially serious limitations’ or both then there is discretion over whether it should be included.

Where there is discretion The health economist will make a decision based on the relative applicability and quality of the available evidence for that question, in discussion with the GDG if required. The ultimate aim is to include studies that are helpful for decision-making in the context of the guideline and the current NHS setting. If several studies are considered of sufficiently high applicability and methodological quality that they could all be included, then the health economist, in discussion with the GDG if required, may decide to include only the most applicable studies and to selectively exclude the remaining studies. All studies excluded on the basis of applicability or methodological limitations will be listed with explanation as excluded economic studies in Appendix K.

The health economist will be guided by the following hierarchies. Setting:  UK NHS  OECD countries with predominantly public health insurance systems (for example, France, Germany, Sweden)  OECD countries with predominantly private health insurance systems (for example, USA, Switzerland)  non-OECD settings (always ‘Not applicable’). Economic study type:  cost–utility analysis  other type of full economic evaluation (cost–benefit analysis, cost-effectiveness analysis, cost–consequences analysis)  comparative cost analysis  non-comparative cost analyses including cost-of-illness studies (always ‘Not applicable’). Year of analysis:  The more recent the study, the more applicable it is. Quality and relevance of effectiveness data used in the economic analysis:  The more closely the effectiveness data used in the economic analysis matches with the outcomes of the studies included in the clinical review the more useful the analysis will be for decision-making in the guideline.

National Clinical Guideline Centre, 2014 82 Type 1 diabetes in adults

Clinical article selection

Appendix D: Clinical article selection

D.1 Diagnosis

Figure 1: Flow chart of clinical article selection for the review of distinguishing between different types of diabetes

Records identified through database searching, n =1968 (+ reruns 1+2 n=556) Additional records identified through Total = 2524 other sources, n = 12

Records screened , n = 2536

Records excluded, n = 2221

Full text articles assessed for eligibility,- n =170 (+ reruns 1+2 n=145) Total = 315

Studies included in review N == 45 (+ reruns 1+2, n=18) Total, = 63 Studies excluded from review, n = 252

Reasons for e xclusion: (see exclusion lists)

National Clinical Guideline Centre, 2014 83 Type 1 diabetes in adults

Clinical article selection

D.2 Education programmes and self-care

Figure 2: Flow chart of clinical article selection for the review of Structured education programmes

Records identified through database searching, n =151 (+ reruns 1+2 n=257) Additional records identified through Total = 408 other sources, n = 12

Records screened , n = 163

Records excluded, n = 102

Full text articles assessed for eligibility,- n =61 (+ reruns 1+2 n=19) Total = 80

Studies included in review n = 13 , (+reruns +2, n=2) , Total N=15 Studies excluded from review, n = 65

Reasons for e xclusion: (see exclusion lists)

National Clinical Guideline Centre, 2014 84 Type 1 diabetes in adults

Clinical article selection

Figure 3: Flow chart of clinical article selection for the review of Carb counting

Records identified through database Additional records identified through searching, n = 646 (+ reruns 1+2, other sources, n = 7 n=109) Total = 755

Records screened, n=762

Records excluded, n =615

Full-text articles assessed for eligibility, n = 39 (+ reruns 1+2, n=8) Total = 47

Studies included in review, n = 11 Studies excluded from review, n= 36 (reruns 1+2, n=0)

National Clinical Guideline Centre, 2014 85 Type 1 diabetes in adults

Clinical article selection

Figure 4: Flow chart of clinical article selection for the review of Glycaemic index diet

Records identified through database Additional records identified through searching, n=331 (+ reruns 1+2, n=22) other sources, n=9 Total = 353

Records screened, n=362

Records excluded, n=295

Full-text articles assessed for eligibility, n=66 (+ reruns 1+2, n=1) Total = 67

Studies included in review, n=5 Studies excluded from review, n=62 (reruns 1+2, n=0) Reasons for exclusion: (see exclusion lists)

National Clinical Guideline Centre, 2014 86 Type 1 diabetes in adults

Clinical article selection

D.3 Blood glucose monitoring

Figure 5: Flow chart of clinical article selection for the review of HbA1c

Records identified through database Additional records identified through searching, N=2613 (+ reruns 1+2, other sources, N=16 n=1582) Total = 4195

Records screened, N=4211

Records excluded, N=3963

Full-text articles assessed for eligibility, N=170 (+reruns 1+2, n=78) Total = 248

Studies included in review, N=30 Studies excluded from review, n =205 (+ reruns 1+2, n=13) Total = 43 Reasons for exclusion: (see exclusion lists)

National Clinical Guideline Centre, 2014 87 Type 1 diabetes in adults

Clinical article selection

Figure 6: Flow chart of clinical article selection for the review of SMBG targets, timing and frequency

Records identified through database Additional records identified through searching, n = 2716, (+ reruns 1+2, other sources, n=0 n=548) Total = 3264)

Records screened, n = 3264

Records excluded, n =3031

Full-text articles assessed for eligibility, n = 207 (+ reruns 1+2, n=26) Total = 233

Studies included in review, n = 44 (+ Studies excluded from review, n= 187 reruns 1+2, n=2) Total = 46 Reasons for exclusion: (see exclusion lists) 2

National Clinical Guideline Centre, 2014 88 Type 1 diabetes in adults

Clinical article selection

Figure 7: Flow chart of clinical article selection for the review of SMBG technologies

Records identified through database Additional records identified through searching, n = 756 (+ reruns 1+2, other sources, n = 1 n=221) Total =977

Records screened, n = 978

Records excluded, n = 829

Full-text articles assessed for eligibility, n = 136 (+ reruns 1+2, n=13) Total = 149

Studies included in review, n = 2 Studies excluded from review, n =147Reasons for exclusion: (see exclusion lists)

National Clinical Guideline Centre, 2014 89 Type 1 diabetes in adults

Clinical article selection

Figure 8: Flow chart of clinical article selection for the review of SMBG versus CGM

Records identified through database Additional records identified through searching, n = 384 (+ reruns 1+2, other sources, n = 8 n=580) Total = 964

Records screened, n = 972

Records excluded, n = 823

Full-text articles assessed for eligibility, n = 115 (+ reruns 1+2, n=34) Total = 149

Studies included in review, n =12 (+ Studies excluded from review, n =135 reruns 1+2, n=2) Total = 14 Reasons for exclusion: (see exclusion lists)

National Clinical Guideline Centre, 2014 90 Type 1 diabetes in adults

Clinical article selection

D.3.1 Insulin therapy

Figure 9: Flow chart of clinical article selection for the review of Rapid-acting insulin

Records identified through database Additional records identified through searching, n = 728) (+ reruns 1+2, other sources, n = 20 cross-ref n=101) Total = 829

Records screened, n =849

Records excluded, n =748

Full-text articles assessed for eligibility, n = 95 (75 + 20) (+ reruns 1+2, n=4) Total =99

Studies included in review, n = 26 (28 Studies excluded from review, n =71 papers) – reruns 1+2, n=0 Reasons for exclusion: (see exclusion lists)

National Clinical Guideline Centre, 2014 91 Type 1 diabetes in adults

Clinical article selection

Figure 10: Flow chart of clinical article selection for the review of Long-acting insulin

Records identified through database Additional records identified through searching, n = 878 (+ reruns 1+2, other sources, n = 9 cross-ref n=231 Total = 1109

Records screened, n = 1118

Records excluded, n = 923

Full-text articles assessed for eligibility, n = 147 (138 + 9) (+ reruns 1+2, n=48) Total = 195

Studies included in review, n = 24 Studies excluded from review, n = studies (26 papers) (+ reruns 1+2, n=2) 167 Total = 28 papers Reasons for exclusion: (see exclusion lists)

National Clinical Guideline Centre, 2014 92 Type 1 diabetes in adults

Clinical article selection

Figure 11: Flow chart of clinical article selection for the review of Mixed insulin

Records identified through database Additional records identified through searching, n = 418 (+ reruns 1+2, other sources, n = 1 cross-ref n=36) = 454

Records screened, n = 455

Records excluded, n = 377

Full-text articles assessed for eligibility, n = 69 (+ reruns 1+2, n=9) Total =78

Studies included in review, n = 14 Studies excluded from review, n =64 (reruns 1+2, n=0) Reasons for exclusion: (see exclusion lists)

National Clinical Guideline Centre, 2014 93 Type 1 diabetes in adults

Clinical article selection

Figure 12: Flow chart of clinical article selection for the review of Metformin

Records identified through database Additional records identified through searching, n = 674 (+ reruns 1+2, other sources, n = 0 n=130) Total = 804

Records screened n = 804

Records excluded n = 732

Full-text articles assessed for eligibility, n = 54 (+ reruns 1+2, n=18) Total = 72

Studies included in review, n =13 (15 Studies excluded from review, n = 54 papers) (+ reruns 1+2, n=3) Reasons for exclusion: (see exclusion Total = 18 papers (16 studies) lists)

National Clinical Guideline Centre, 2014 94 Type 1 diabetes in adults

Clinical article selection

Figure 13: Flow chart of clinical article selection for the review of Once versus twice daily basal insulin

Records identified through database Additional records identified through searching, n = 250 other sources, n = 1

Records screened, n = 251

Records excluded, n = 234

Full-text articles assessed for eligibility, n = 17

Studies included in review, n =1 Studies excluded from review, n =16

Reasons for exclusion: (see exclusion lists)

National Clinical Guideline Centre, 2014 95 Type 1 diabetes in adults

Clinical article selection

Figure 14: Flow chart of clinical article selection for the review of Insulin delivery (needle length, site and rotation)

Records identified through database Additional records identified through searching, n =90 (+ reruns 1+2, other sources, n = 0 n=112) Total = 202

Records screened, n = 202

Records excluded, n =140

Full-text articles assessed for eligibility, n = 51 (+ reruns 1+2, n=11) Total = 62

Studies included in review, n =12 Studies excluded from review, n =48 (+ reruns 1+2, n=2) Total = 14 Reasons for exclusion: (see exclusion lists)

National Clinical Guideline Centre, 2014 96 Type 1 diabetes in adults

Clinical article selection

Figure 15: Flow chart of clinical article selection for the review of Insulin delivery (needle length, site and rotation)

Records identified through database Additional records identified through searching, n =90 (+ reruns 1+2, other sources, n = 0 n=112) Total = 202

Records screened, n = 202

Records excluded, n =140

Full-text articles assessed for eligibility, n = 51 (+ reruns 1+2, n=11) Total = 62

Studies included in review, n =12 Studies excluded from review, n =48 (+ reruns 1+2, n=2) Total = 14 Reasons for exclusion: (see exclusion lists)

National Clinical Guideline Centre, 2014 97 Type 1 diabetes in adults

Clinical article selection

D.4 Hypoglycaemia

Figure 16: Flow chart of clinical article selection for the review of the Identification of hypoglycaemia awareness

Records identified through database Additional records identified through searching, n=1535 (+ reruns 1+2, other sources, n=7 n=293) Total = 1828

Records screened, n=1835

Records excluded, n=1792

Full-text articles assessed for eligibility, n=36 (+ reruns 1+2, n=7). Total = 43

Studies included in review, n=12 (+ Studies excluded from review, n=29 reruns 1+2, n=2) Total = 14 (+7 relevant conference abstracts)

National Clinical Guideline Centre, 2014 98 Type 1 diabetes in adults

Clinical article selection

Figure 17: Flow chart of clinical article selection for the review of Strategies to recover hypoglycaemia awareness

Records identified through database Additional records identified through searching, n=1535 (+ reruns 1+2, other sources, n=6 n=293) Total = 1828

Records screened, n=1834

Records excluded, n=1704

Full-text articles assessed for eligibility, n=96 (+ reruns 1+2, n=34) Total = 130

Studies included in review, n=19 (+ Studies excluded from review, n=110 reruns 1+2, n=1) Total = 20

National Clinical Guideline Centre, 2014 99 Type 1 diabetes in adults

Clinical article selection

D.5 Ketone monitoring

Figure 18: Flow chart of clinical article selection for the review of Ketone self-monitoring and in- hospital self-monitoring

Records identified through database Additional records identified through searching, n = 54 (+ reruns 1+2, other sources, n = 4 n=909) Total = 963

Records screened, n = 967

Records excluded, n = 908

Full-text articles assessed for eligibility, n = 30 (+ reruns 1+2, n=29) Total = 59

Studies included in review, n = 5 Studies excluded from review, n =253 (+ reruns 1+2, n=1) Total = 6 Reasons for exclusion: (see exclusion lists)

National Clinical Guideline Centre, 2014 100 Type 1 diabetes in adults

Clinical article selection

D.6 Arterial risk control

Figure 19: Flow chart of clinical article selection for the review of Aspirin

Records identified through database Additional records identified through searching, n = 166 (+ reruns 1+2, other sources, n = 7 n=334) Total = 500

Records screened, n = 507

Records excluded, n = 470

Full-text articles assessed for eligibility, n = 29 (+ reruns 1+2, n=8) Total = 37

Studies included in review, n = 2 Studies excluded from review, n =35 (reruns 1+2, n=0) Reasons for exclusion: (see exclusion lists)

National Clinical Guideline Centre, 2014 101 Type 1 diabetes in adults

Clinical article selection

D.7 Inpatient management

Figure 20: Flow chart of clinical article selection for the review of IV insulin

Records identified through database Additional records identified through searching, n = 719 (+ reruns 1+2, other sources, n = 1 n=73) Total = 792

Records screened, n = 793

Records excluded, n = 693

Full-text articles assessed for eligibility, n = 94 (+ reruns 1+2, n=6) Total = 100

Studies included in review, n = 6 Studies excluded from review, n = 94 (reruns 1+2, n=0)

National Clinical Guideline Centre, 2014 102 Type 1 diabetes in adults

Clinical article selection

D.8 Complications

Figure 21: Flow chart of clinical article selection for the review of Gastroparesis

Records identified through database Additional records identified through searching, n = 719 (+ reruns 1+2, other sources, n = 2 (old 2004 GL) n=413) Total = 1132 n=19 (cross-referencing)

Records screened, n =1151

Records excluded, n=1011

Full-text articles assessed for eligibility, n = 95 (+ reruns 1+2, n=45) Total = 140

Studies included in review, n = 13 Studies excluded from review, n=122 (+ reruns 1+2, n=5) Total = 18 Reasons for exclusion: (see exclusion lists)

National Clinical Guideline Centre, 2014 103 Type 1 diabetes in adults

Clinical article selection

Figure 22: Flow chart of clinical article selection for the review of Acute painful neuropathy

Records identified through database Additional records identified through searching, n = 1783 (+ reruns 1+2, other sources, n = 30 n=212) Total = 1894

Records screened, n = 1924

Records excluded, n = 1830

Full-text articles assessed for eligibility, n = 86 (+ reruns 1+2, n=8) Total = 94

Studies included in review, n = 1 Studies excluded from review, n = 93

National Clinical Guideline Centre, 2014 104 Type 1 diabetes in adults

Clinical article selection

Figure 23: Flow chart of clinical article selection for the review of Thyroid disease - frequency of monitoring

Records identified through database Additional records identified through searching, n = 954 (+ reruns 1+2, other sources, n=0 n=82) Total = 1036

Records screened, n = 1036

Records excluded, n = 819

Full-text articles assessed for eligibility, n = 211 (+ reruns 1+2, n=6) Total = 217

Studies included in review, n =22 Studies excluded from review, n =195

National Clinical Guideline Centre, 2014 105 Type 1 diabetes in adults

Economic article selection

Appendix E: Economic article selection

Figure 24: Flow chart of clinical article selection for the guideline

Records identified through database Additional records identified through searching, n=1412 other sources, n=0

Records screened in 1st sift, n=1412

Records excluded* in 1st sift, n=1213

Full-text articles assessed for eligibility in 2nd sift, n=199

Records excluded* in 2nd sift, n=166

Full-text articles assessed for applicability and quality of methodology, n=33

Studies included, n=14 Studies selectively Studies excluded, n=0 excluded, n=19

Reasons for exclusion: see Reasons for exclusion: see Appendix M Appendix M

* Non-relevant population, intervention, comparison, design or setting; non-English language

National Clinical Guideline Centre, 2014 106 Type 1 diabetes in adults

Literature search strategies

Appendix F: Literature search strategies Contents Introduction Search methodology Section Error! Study filter terms eference source not found. Error! Systematic reviews (SR) eference source not found. Error! Randomized controlled trials (RCT) eference source not found. Error! Observational studies (OBS) eference source not found. Error! Cohort studies (COHORT) eference source not found. Error! Prognostic studies (PROG) eference source not found. Error! Economic studies (ECON) eference source not found. Error! Quality of life studies (QOL) eference source not found. Section Error! Standard population search strategy eference This population was used for all search questions unless stated source not found. Section Error! Searches for specific questions with intervention (and population where eference different from A.2) source not found. Error! Differential diagnosis eference source not

National Clinical Guideline Centre, 2014 107 Type 1 diabetes in adults

Literature search strategies

found. Error! Structured education programmes eference source not found. Error! Carbohydrates eference source not found. Error! Diet eference source not found.

Error! HbA1c eference source not found. Error! Self-monitoring of blood glucose (SMBG) eference source not found. Error! Glucose monitoring eference source not found. Error! Technologies eference source not found. Error! Long-acting/basal insulins eference source not found. Error! Rapid-acting insulins eference source not found. Error! Mixed insulins eference source not found. Error! Metformin and GLP-1 agonists eference source not found. Error! Glucose injections eference source not found.

National Clinical Guideline Centre, 2014 108 Type 1 diabetes in adults

Literature search strategies

Error! Pancreas transplantation eference source not found. Error! Hypoglycaemic awareness eference source not found. Error! Ketone monitoring eference source not found. Error! Aspirin for prevention of cardiovascular events eference source not found. Error! IV insulin eference source not found. Error! Gastroparesis eference source not found. Error! Thyroid disease eference source not found. Error! Neuropathy eference source not found. Section Error! Economic searches eference source not found. Error! Economic reviews eference source not found. F.4.2 Quality of life reviews Introduction

Search strategies used for the Type 1 Diabetes guideline update were run in accordance with the NICE Guidelines Manual 20121. All searches were run up to 28 August 2014 unless otherwise stated. Any studies added to the databases after this date were not included unless specifically stated in the text. Where possible searches were limited to retrieve material published in English.

National Clinical Guideline Centre, 2014 109 Type 1 diabetes in adults

Literature search strategies

For questions not covered in the original NICE guideline the dates searched for each database were as follows in Table 1. For questions in the original guideline that were updated, searches were undertaken from 2003 onwards – the date of the last searches conducted for the original guideline. Dates searched for each question are documented in the tables under each question in section F.3Error! Reference source not found..

Table 1: Database date parameters Database Dates searched Medline 1946 – 28 August 2014 Embase 1980 – August 2014 (week 34) The Cochrane Library Cochrane Reviews to 2014 Issue 7 of 12 CENTRAL to 2014 Issue 7 of 12 DARE, HTA and NHSEED to 2013 Issue 3 of 4

Clinical searches

Searches for clinical reviews were run in Medline (OVID), Embase (OVID) and the Cochrane Library (Wiley). Typically, searches were constructed in the following way:  A PICO format was used for intervention searches. Population (P) terms were combined with Intervention (I) and sometimes Comparison (C) terms (as indicated in the tables under each individual question in F.3). An intervention can be a drug, a procedure or a diagnostic test. Outcomes (O) are rarely used in search strategies for interventions. Study type filters were added where appropriate (see F.1).  A PEO format was used for prognosis searches where population (P) terms were combined with exposure (E) terms and sometimes outcomes (O).

In addition to the databases outlined above, searches for the structured education question (F.3.1) were run in PsycINFO (OVID). Economic searches

Searches for economic evidence were run in Medline (Ovid), Embase (Ovid), the NHS Economic Evaluations Database (NHS EED), the Health Technology Assessment (HTA) database and the Health Economic Evaluation Database (HEED). NHS EED and HTA were searched via the Cochrane (Wiley) or Centre for Reviews and Dissemination (CRD) interfaces. For Medline and Embase an economic filter (see F.1.6) was added to the standard populations (see F.2). All other searches were conducted using only population terms.

F.1 Study filter search terms

F.1.1 Systematic review (SR) search terms

Medline search terms 1. meta-analysis/ 2. meta-analysis as topic/ 3. (meta analy* or metanaly* or metaanaly*).ti,ab. 4. ((systematic* or evidence*) adj2 (review* or overview*)).ti,ab. 5. (reference list* or bibliograph* or hand search* or manual search* or relevant journals).ab. 6. (search strategy or search criteria or systematic search or study selection or data extraction).ab. 7. (search* adj4 literature).ab.

National Clinical Guideline Centre, 2014 110 Type 1 diabetes in adults

Literature search strategies

8. (medline or pubmed or cochrane or embase or psychlit or psyclit or psychinfo or psycinfo or cinahl or science citation index or bids or cancerlit).ab. 9. cochrane.jw. 10. or/1-9

Embase search terms 1. systematic review/ 2. meta-analysis/ 3. (meta analy* or metanaly* or metaanaly*).ti,ab. 4. ((systematic or evidence) adj2 (review* or overview*)).ti,ab. 5. (reference list* or bibliograph* or hand search* or manual search* or relevant journals).ab. 6. (search strategy or search criteria or systematic search or study selection or data extraction).ab. 7. (search* adj4 literature).ab. 8. (medline or pubmed or cochrane or embase or psychlit or psyclit or psychinfo or psycinfo or cinahl or science citation index or bids or cancerlit).ab. 9. ((pool* or combined) adj2 (data or trials or studies or results)).ab. 10. cochrane.jw. 11. or/1-10

F.1.2 Randomised controlled trials (RCT) search terms

Medline search terms 1. randomized controlled trial.pt. 2. controlled clinical trial.pt. 3. randomi#ed.ab. 4. placebo.ab. 5. randomly.ab. 6. clinical trials as topic.sh. 7. trial*.ti. 8. or/1-7

Embase search terms 1. random*.ti,ab. 2. factorial*.ti,ab. 3. (crossover* or cross over*).ti,ab. 4. ((doubl* or singl*) adj blind*).ti,ab. 5. (assign* or allocat* or volunteer* or placebo*).ti,ab. 6. crossover procedure/ 7. single blind procedure/ 8. randomized controlled trial/ 9. double blind procedure/ 10. or/1-9

F.1.3 Observational studies (OBS) search terms

Medline search terms 1. epidemiologic studies/

National Clinical Guideline Centre, 2014 111 Type 1 diabetes in adults

Literature search strategies

2. exp case control studies/ 3. exp cohort studies/ 4. cross-sectional studies/ 5. case control.ti,ab. 6. (cohort adj (study or studies or analys*)).ti,ab. 7. ((follow up or observational or uncontrolled or non randomi#ed or nonrandomi#ed or epidemiologic*) adj (study or studies)).ti,ab. 8. ((longitudinal or retrospective or prospective or cross sectional) and (study or studies or review or analys* or cohort*)).ti,ab. 9. or/1-8

Embase search terms 1. clinical study/ 2. exp case control study/ 3. family study/ 4. longitudinal study/ 5. retrospective study/ 6. prospective study/ 7. cross-sectional study/ 8. cohort analysis/ 9. follow-up/ 10. cohort*.ti,ab. 11. 9 and 10 12. case control.ti,ab. 13. (cohort adj (study or studies or analys*)).ti,ab. 14. ((follow up or observational or uncontrolled or non randomi#ed or nonrandomi#ed or epidemiologic*) adj (study or studies)).ti,ab. 15. ((longitudinal or retrospective or prospective or cross sectional) and (study or studies or review or analys* or cohort*)).ti,ab. 16. or/1-8,11-15

F.1.4 Cohort studies search terms

Medline search terms 1. exp cohort studies/ 2. cross-sectional studies/ 3. retrospective studies/ 4. ((prospective or cross sectional or retrospective or follow up or longitudinal or comparative) and (study or studies or review or analys*)).ti,ab. 5. comparative study.pt. 6. (cohort* or participant*).ti,ab. 7. or/1-6

Embase search terms 1. comparative study/ 2. longitudinal study/ 3. prospective study/ 4. cross-sectional study/

National Clinical Guideline Centre, 2014 112 Type 1 diabetes in adults

Literature search strategies

5. retrospective study/ 6. cohort analysis/ 7. ((prospective or retrospective or cross sectional or follow up or longitudinal or comparative) and (study or studies or review or analys*)).ti,ab. 8. (cohort* or participant*).ti,ab. 9. or/1-8

Cochrane search terms #1. MeSH descriptor: [cohort studies] explode all trees #2. MeSH descriptor: [cross-sectional studies] this term only #3. ((prospective or retrospective or “cross sectional” or “follow up” or longitudinal or comparative) and (study or studies or review or analys*)):ti,ab #4. comparative study:pt #5. (cohort* or participant*):ti,ab #6. #1 or #2 or #3 or #4 or #5

F.1.5 Prognostic (PROG) studies search terms

Medline search terms 1. predict.ti. 2. (validat* or rule*).ti,ab. 3. (predict* and (outcome* or risk* or model*)).ti,ab. 4. ((history or variable* or criteria or scor* or characteristic* or finding* or factor*) and (predict* or model* or decision* or identif* or prognos*)).ti,ab. 5. decision*.ti,ab. and logistic models/ 6. (decision* and (model* or clinical*)).ti,ab. 7. (prognostic and (history or variable* or criteria or scor* or characteristic* or finding* or factor* or model*)).ti,ab. 8. (stratification or discrimination or discriminate or c statistic or "area under the curve" or auc or calibration or indices or algorithm or multivariable).ti,ab. 9. roc curve/ 10. or/1-9

Embase search terms 1. predict.ti. 2. (validat* or rule*).ti,ab. 3. (predict* and (outcome* or risk* or model*)).ti,ab. 4. ((history or variable* or criteria or scor* or characteristic* or finding* or factor*) and (predict* or model* or decision* or identif* or prognos*)).ti,ab. 5. decision*.ti,ab. and statistical model/ 6. (decision* and (model* or clinical*)).ti,ab. 7. (prognostic and (history or variable* or criteria or scor* or characteristic* or finding* or factor* or model*)).ti,ab. 8. (stratification or discrimination or discriminate or c statistic or "area under the curve" or auc or calibration or indices or algorithm or multivariable).ti,ab. 9. receiver operating characteristic/ 10. or/1-9

National Clinical Guideline Centre, 2014 113 Type 1 diabetes in adults

Literature search strategies

F.1.6 Health economic (ECON) search terms

Medline search terms 1. economics/ 2. value of life/ 3. exp "costs and cost analysis"/ 4. exp economics, hospital/ 5. exp economics, medical/ 6. exp resource allocation/ 7. economics, nursing/ 8. economics, pharmaceutical/ 9. exp "fees and charges"/ 10. exp budgets/ 11. budget*.ti,ab. 12. cost*.ti,ab. 13. (economic* or pharmaco?economic*).ti,ab. 14. (price* or pricing*).ti,ab. 15. (financ* or fee or fees or expenditure* or saving*).ti,ab. 16. (value adj2 (money or monetary)).ti,ab. 17. resourc* allocat*.ti,ab. 18. (fund or funds or funding* or funded).ti,ab. 19. (ration or rations or rationing* or rationed).ti,ab. 20. ec.fs. 21. or/1-20

Embase search terms 1. health economics/ 2. exp economic evaluation/ 3. exp health care cost/ 4. exp fee/ 5. budget/ 6. funding/ 7. resource allocation/ 8. budget*.ti,ab. 9. cost*.ti,ab. 10. (economic* or pharmaco?economic*).ti,ab. 11. (price* or pricing*).ti,ab. 12. (financ* or fee or fees or expenditure* or saving*).ti,ab. 13. (value adj2 (money or monetary)).ti,ab. 14. resourc* allocat*.ti,ab. 15. (fund or funds or funding* or funded).ti,ab. 16. (ration or rations or rationing* or rationed).ti,ab. 17. or/1-16

F.1.7 Quality of life (QoL) search terms

Medline search terms

National Clinical Guideline Centre, 2014 114 Type 1 diabetes in adults

Literature search strategies

1. quality-adjusted life years/ 2. sickness impact profile/ 3. (quality adj2 (wellbeing or well being)).ti,ab. 4. sickness impact profile.ti,ab. 5. disability adjusted life.ti,ab. 6. (qal* or qtime* or qwb* or daly*).ti,ab. 7. (euroqol* or eq5d* or eq 5d*).ti,ab. 8. (qol* or hql* or hqol* or h qol* or hrqol* or hr qol*).ti,ab. 9. (health utility* or utility score* or disutilit*).ti,ab. 10. (hui or hui1 or hui2 or hui3).ti,ab. 11. health* year* equivalent*.ti,ab. 12. (hye or hyes).ti,ab. 13. rosser.ti,ab. 14. (willingness to pay or time tradeoff or time trade off or tto or standard gamble*).ti,ab. 15. (sf36 or sf 36 or short form 36 or shortform 36 or shortform36).ti,ab. 16. (sf20 or sf 20 or short form 20 or shortform 20 or shortform20).ti,ab. 17. (sf12 or sf 12 or short form 12 or shortform 12 or shortform12).ti,ab. 18. (sf8 or sf 8 or short form 8 or shortform 8 or shortform8).ti,ab. 19. (sf6 or sf 6 or short form 6 or shortform 6 or shortform6).ti,ab. 20. or/1-20

Embase search terms 1. quality adjusted life year/ 2. "quality of life index"/ 3. short form 12/ or short form 20/ or short form 36/ or short form 8/ 4. sickness impact profile/ 5. (quality adj2 (wellbeing or well being)).ti,ab. 6. sickness impact profile.ti,ab. 7. disability adjusted life.ti,ab. 8. (qal* or qtime* or qwb* or daly*).ti,ab. 9. (euroqol* or eq5d* or eq 5d*).ti,ab. 10. (qol* or hql* or hqol* or h qol* or hrqol* or hr qol*).ti,ab. 11. (health utility* or utility score* or disutilit*).ti,ab. 12. (hui or hui1 or hui2 or hui3).ti,ab. 13. health* year* equivalent*.ti,ab. 14. (hye or hyes).ti,ab. 15. rosser.ti,ab. 16. (willingness to pay or time tradeoff or time trade off or tto or standard gamble*).ti,ab. 17. (sf36 or sf 36 or short form 36 or shortform 36 or shortform36).ti,ab. 18. (sf20 or sf 20 or short form 20 or shortform 20 or shortform20).ti,ab. 19. (sf12 or sf 12 or short form 12 or shortform 12 or shortform12).ti,ab. 20. (sf8 or sf 8 or short form 8 or shortform 8 or shortform8).ti,ab. 21. (sf6 or sf 6 or short form 6 or shortform 6 or shortform6).ti,ab. 22. or/1-21

National Clinical Guideline Centre, 2014 115 Type 1 diabetes in adults

Literature search strategies

F.2 Population search strategies

Medline search terms 1. diabetes mellitus, type 1/ 2. diabetic ketoacidosis/ 3. ((diabet* or dm) adj4 (type 1 or type1 or type i or type one)).ti,ab. 4. (diabet* adj2 (autoimmun* or auto immun*)).ti,ab. 5. lada.ti,ab. 6. (diabet* adj2 (brittle or labile)).ti,ab. 7. (diabet* adj2 (sudden onset or juvenile or childhood)).ti,ab. 8. (diabet* adj3 (keto* or acido* or gastropare*)).ti,ab. 9. (dm1 or iddm or t1d* or dka).ti,ab. 10. ((diabet* adj2 (insulin depend* or insulin deficien*)) not non insulin depend*).ti,ab. 11. diabetes mellitus.ti. 12. (diabet* adj3 (type 2 or type ii)).ti. 13. 11 not 12 14. or/1-10,13 15. ((children or adolescen* or school* or infant* or teenage* or paediatric* or pediatric*) not (adult* or onset)).ti. 16. (pregnan* or gestation*).ti. 17. 14 not (15 or 16) 18. letter/ 19. editorial/ 20. news/ 21. exp historical article/ 22. anecdotes as topic/ 23. comment/ 24. case report/ 25. (letter or comment*).ti. 26. or/18-25 27. 26 not (randomized controlled trial/ or random*.ti,ab.) 28. animals/ not humans/ 29. exp animals, laboratory/ 30. exp animal experimentation/ 31. exp models, animal/ 32. exp rodentia/ 33. (rat or rats or mouse or mice).ti. 34. or/27-33 35. 17 not 34

Embase search terms 1. insulin dependent diabetes mellitus/ 2. juvenile diabetes mellitus/ 3. diabetic ketoacidosis/ 4. ((diabet* or dm) adj4 (type 1 or type1 or type i or type one)).ti,ab. 5. (diabet* adj2 (autoimmun* or auto immun*)).ti,ab.

National Clinical Guideline Centre, 2014 116 Type 1 diabetes in adults

Literature search strategies

6. lada.ti,ab. 7. (diabet* adj2 (brittle or labile)).ti,ab. 8. (diabet* adj2 (sudden onset or juvenile or childhood)).ti,ab. 9. (diabet* adj3 (keto* or acido* or gastropare*)).ti,ab. 10. (dm1 or iddm or t1d* or dka).ti,ab. 11. ((diabet* adj2 (insulin depend* or insulin deficien*)) not non insulin depend*).ti,ab. 12. diabetes mellitus.ti. 13. (diabet* adj3 (type 2 or type ii)).ti. 14. 12 not 13 15. or/1-11,14 16. ((children or adolescen* or school* or infant* or teenage* or paediatric* or pediatric*) not (adult* or onset)).ti. 17. (pregnan* or gestation*).ti. 18. 15 not (16 or 17) 19. letter.pt. or letter/ 20. note.pt. 21. editorial.pt. 22. case report/ or case study/ 23. (letter or comment*).ti. 24. or/19-23 25. 24 not (randomized controlled trial/ or random*.ti,ab.) 26. animal/ not human/ 27. nonhuman/ 28. exp animal experiment/ 29. exp experimental animal/ 30. animal model/ 31. exp rodent/ 32. (rat or rats or mouse or mice).ti. 33. or/25-32 34. 18 not 33

Cochrane search terms #1. MeSH descriptor diabetes mellitus, type 1 explode all trees #2. MeSH descriptor diabetic ketoacidosis, this term only #3. ((diabet* or dm) near/4 ("type 1" or type1 or "type i" or "type one")):ti,ab #4. (diabet* near/2 (autoimmun* or "auto immun*")):ti,ab #5. (diabet* near/2 (brittle or labile)):ti,ab #6. (diabet* near/2 ("sudden onset" or juvenile or child*)):ti,ab #7. (diabet* near/3 (keto* or acido* or gastropare*)):ti,ab #8. (dm1 or iddm or t1d* or dka or lada):ti,ab #9. (diabet* near/2 (insulin next depend*)):ti,ab #10. (#9 and not "non insulin dependent") #11. (#1 or #2 or #3 or #4 or #5 or #6 or #7 or #8 or #10) #12. "diabetes mellitus":ti #13. (diabet* near/3 ("type 2" or "type ii")):ti #14. (#12 and not #13)

National Clinical Guideline Centre, 2014 117 Type 1 diabetes in adults

Literature search strategies

#15. (#11 or #14) #16. ((children or adolescen* or school* or infant* or teenage* or paediatric* or pediatric*) not (adult* or onset)):ti #17. (#15 and not #16) #18. (pregnan* or gestation*):ti #19. (#17 and not #18)

PsycINFO search terms 1. ((diabet* or dm) adj4 (type 1 or type1 or type i or type one)).ti,ab. 2. (diabet* adj2 (autoimmun* or auto immun*)).ti,ab. 3. lada.ti,ab. 4. (diabet* adj2 (brittle or labile)).ti,ab. 5. (diabet* adj2 (sudden onset or juvenile or childhood)).ti,ab. 6. (diabet* adj3 (keto* or acido* or gastropare*)).ti,ab. 7. (dm1 or iddm or t1d* or dka).ti,ab. 8. ((diabet* adj2 (insulin depend* or insulin deficien*)) not non insulin depend*).ti,ab. 9. diabetes mellitus.ti. 10. (diabet* adj3 (type 2 or type ii)).ti. 11. 9 not 10 12. or/1-8,11 13. ((children or adolescen* or school* or infant* or teenage* or paediatric* or pediatric*) not (adult* or onset)).ti. 14. 12 not 13

F.3 Searches for specific questions

F.3.1 Differential diagnosis

For the following question a broad diabetes population was used, hence the searches for this question are reproduced in full below. 1. In adults and young people with diabetes, what is the best marker (c-peptides plus or minus antibodies) to distinguish between type 1 diabetes, type 2 diabetes and other forms of diabetes?

Search constructed by combining the columns in the following table using the AND Boolean operator Population Intervention Study filter used Date parameters Type 1 or type 2 Markers The following filters 2003 - 28 August 2014 diabetes were used in Medline [see Table 1] and Embase only: RCTs or SRs or cohort studies [Medline and Embase only]

Medline search terms 1. diabetes mellitus, type 1/ 2. diabetic ketoacidosis/ 3. ((diabet* or dm) adj4 (type 1 or type1 or type i or type one)).ti,ab. 4. (diabet* adj2 (autoimmun* or auto immun*)).ti,ab. 5. (lada or mody).ti,ab. 6. (diabet* adj2 (brittle or labile)).ti,ab.

National Clinical Guideline Centre, 2014 118 Type 1 diabetes in adults

Literature search strategies

7. (diabet* adj2 (sudden onset or maturity onset or juvenile or childhood)).ti,ab. 8. (diabet* adj3 (keto* or acido* or gastropare*)).ti,ab. 9. (dm1 or iddm or t1d* or dka).ti,ab. 10. ((diabet* adj2 (insulin depend* or insulin deficien*)) not non insulin depend*).ti,ab. 11. diabetes mellitus.ti. 12. or/1-11 13. (pregnan* or gestation*).ti. 14. 12 not 13 15. letter/ 16. editorial/ 17. news/ 18. exp historical article/ 19. anecdotes as topic/ 20. comment/ 21. case report/ 22. (letter or comment*).ti. 23. or/15-22 24. 23 not (randomized controlled trial/ or random*.ti,ab.) 25. animals/ not humans/ 26. exp animals, laboratory/ 27. exp animal experimentation/ 28. exp models, animal/ 29. exp rodentia/ 30. (rat or rats or mouse or mice).ti. 31. or/24-30 32. 14 not 31 33. c-peptide/ 34. *autoantibodies/ 35. glutamate decarboxylase/ 36. insulinoma/ 37. glucose-6-phosphatase/ 38. c peptide*.ti,ab. 39. ((islet cell or decarboxylase or glutamic or insulinoma) and (antibod* or anti bod*or autoantibod*)).ti,ab. 40. zinc transporter 8.ti,ab. 41. (islet adj5 (phosphatase or catalytic)).ti,ab. 42. (igrp* or ica* or ia-2* or ia2* or znt8* or gad*).ti,ab. 43. or/33-42 44. 32 and 43 45. (diagnos* or screen* or test*).ti,ab,hw. 46. exp "sensitivity and specificity"/ 47. roc curve/ 48. area under curve/ 49. proportional hazards models/ 50. (roc or auc or (area and curve)).ti,ab.

National Clinical Guideline Centre, 2014 119 Type 1 diabetes in adults

Literature search strategies

51. (sensitivity or specificity).ti,ab. 52. gold standard.ab. 53. (predictive value* or ppv or npv).ti,ab. 54. likelihood ratio*.ti,ab. 55. or/46-55 56. 44 and 55

Embase search terms 1. insulin dependent diabetes mellitus/ 2. juvenile diabetes mellitus/ 3. diabetic ketoacidosis/ 4. ((diabet* or dm) adj4 (type 1 or type1 or type i or type one)).ti,ab. 5. (diabet* adj2 (autoimmun* or auto immun*)).ti,ab. 6. (lada or mody).ti,ab. 7. (diabet* adj2 (brittle or labile)).ti,ab. 8. (diabet* adj2 (sudden onset or maturity onset or juvenile or childhood)).ti,ab. 9. (diabet* adj3 (keto* or acido* or gastropare*)).ti,ab. 10. (dm1 or iddm or t1d* or dka).ti,ab. 11. ((diabet* adj2 (insulin depend* or insulin deficien*)) not non insulin depend*).ti,ab. 12. diabetes mellitus.ti. 13. or/1-12 14. (pregnan* or gestation*).ti. 15. 13 not 14 16. letter.pt. or letter/ 17. note.pt. 18. editorial.pt. 19. case report/ or case study/ 20. (letter or comment*).ti. 21. or/16-20 22. 21 not (randomized controlled trial/ or random*.ti,ab.) 23. animal/ not human/ 24. nonhuman/ 25. exp animal experiment/ 26. exp experimental animal/ 27. animal model/ 28. exp rodent/ 29. (rat or rats or mouse or mice).ti. 30. or/22-29 31. 15 not 30 32. c peptide/ 33. glutamate decarboxylase 65 antibody/ 34. insulinoma/ 35. *autoantibody/ 36. glucose 6 phosphatase/ 37. ((islet cell or decarboxylase or glutamic or insulinoma) and (antibod* or anti bod*or autoantibod*)).ti,ab.

National Clinical Guideline Centre, 2014 120 Type 1 diabetes in adults

Literature search strategies

38. zinc transporter 8.ti,ab. 39. (islet adj5 (phosphatase or catalytic)).ti,ab. 40. (igrp* or ica* or ia-2* or ia2* or znt8* or gad*).ti,ab. 41. or/32-40 42. 31 and 41 43. (diagnos* or screen* or test*).ti,ab,hw. 44. exp "sensitivity and specificity"/ 45. receiver operating characteristic/ 46. area under the curve/ 47. proportional hazards model/ 48. (roc or auc or (area and curve)).ti,ab. 49. (sensitivity or specificity).ti,ab. 50. gold standard.ab. 51. (predictive value* or ppv or npv).ti,ab. 52. likelihood ratio*.ti,ab. 53. or/45-54 54. 42 and 53

Cochrane search terms #1. MeSH descriptor: [diabetes mellitus, type 1] explode all trees #2. MeSH descriptor: [diabetic ketoacidosis] this term only #3. ((diabet* or dm) near/4 ("type 1" or type1 or "type i" or "type one")):ti,ab #4. (diabet* near/2 (autoimmun* or "auto immun*")):ti,ab #5. (diabet* near/2 (brittle or labile)):ti,ab #6. (diabet* near/2 ("sudden onset" or "maturity onset" or juvenile or child*)):ti,ab #7. (diabet* near/3 (keto* or acido* or gastropare*)):ti,ab #8. (dm1 or iddm or t1d* or dka or lada or mody):ti,ab #9. (diabet* near/2 (insulin next depend*)):ti,ab #10. #9 and not "non insulin dependent" #11. diabetes mellitus:ti #12. #1 or #2 or #3 or #4 or #5 or #6 or #7 or #8 or #9 or #10 or #11 #13. (pregnan* or gestation*):ti #14. #12 not #13 #15. MeSH descriptor: [c-peptide] this term only #16. MeSH descriptor: [autoantibodies] this term only #17. MeSH descriptor: [glutamate decarboxylase] this term only #18. MeSH descriptor: [insulinoma] this term only #19. MeSH descriptor: [glucose-6-phosphatase] this term only #20. (c next (peptide or peptides)):ti,ab #21. ((“islet cell” or decarboxylase or glutamic or insulinoma) and (antibody or antibodies or “anti body” or “anti bodies” or autoantibody or autoantibodies)):ti,ab #22. (“zinc transporter 8”):ti,ab #23. (islet* and (phosphatase or catalytic)):ti,ab #24. (igrp* or ica* or “ia-2” or ia2* or znt8* or gad*):ti,ab #25. #15 or #16 or #17 or #18 or #19 or #20 or #21 or #22 or #23 or #24 #26. #14 and #25

National Clinical Guideline Centre, 2014 121 Type 1 diabetes in adults

Literature search strategies

#27. MeSH descriptor: [diagnosis] explode all trees #28. (diagnos* or screen* or test*):ti,ab #29. MeSH descriptor: [sensitivity and specificity] explode all trees #30. MeSH descriptor: [area under curve] this term only #31. MeSH descriptor: [proportional hazards models] this term only #32. ((roc or auc) or (area and curve)):ti,ab #33. (sensitivity or specificity):ti,ab #34. “gold standard”:ab #35. (“predictive value” or ppv or npv):ti,ab #36. “likelihood ratio”:ti,ab #37. #27 or #28 or #29 or #30 or #31 or #32 or #33 or #34 or #35 or #36 #38. #37 and #26

F.3.2 Structured education programmes 2. In adults with type 1 diabetes, what is the most effective structured education programme?

Search constructed by combining the columns in the following table using the AND Boolean operator Population Intervention Study filter used Date parameters Type 1 diabetes Structured education The following filters 2003 – 28 August 2014 programme were used in Medline [see Table 1] and Embase only: RCT, SR

Medline search terms 1. self care/ 2. patient education as topic/ 3. (self adj (manage* or care or regulat* or monitor*)).ti. 4. (educat* or learn* or teach* or train* or program* or dafne).ti. 5. ((train* or teach* or educat* or psycho*) and (model* or program* or structured or intervention* or support or diet* or eat*)).ab. 6. or/1-5

Embase search terms 1. patient education/ 2. self care/ 3. education program/ 4. eating/ 5. education/ 6. learning/ 7. diabetes education/ 8. (self adj (manage* or care or regulat* or monitor*)).ti. 9. (educat* or learn* or teach* or train* or program* or dafne).ti. 10. ((train* or teach* or educat* or psycho*) and (model* or program* or structured or intervention* or support or diet* or eat*)).ab. 11. or/1-10

PsycINFO search terms 1. health education/

National Clinical Guideline Centre, 2014 122 Type 1 diabetes in adults

Literature search strategies

2. client education/ 3. self management/ 4. educational programs/ 5. individual education programs/ 6. (self adj (manage* or care or regulat* or monitor*)).ti. 7. (educat* or learn* or teach* or train* or program* or dafne).ti. 8. ((train* or teach* or educat* or psycho*) and (model* or program* or structured or intervention* or support or diet* or eat*)).ab. 9. or/1-8

Cochrane search terms #1. MeSH descriptor patient education as topic, this term only #2. MeSH descriptor self care, this term only #3. (self next (manage* or care or regulat* or monitor*)):ti #4. (educat* or learn* or teach* or train* or program* or dafne):ti #5. ((train* or teach* or educat* or psycho*) and (model* or program* or structured or intervention* or support or diet* or eat*)):ab #6. (#1 or #2 or #3 or #4 or #5)

F.3.3 Carbohydrates 3. In adults with type 1 diabetes, what is the clinical and cost-effectiveness of carbohydrate counting/restriction for optimal diabetic control?

Search constructed by combining the columns in the following table using the AND Boolean operator Population Intervention Study filter used Date parameters Type 1 diabetes Carbohydrates The following filters All available dates [see were used in Medline Table 1] and Embase only: RCTs or SRs or cohort studies [Medline and Embase only]

Medline search terms 1. (carb or carbs or carbohydrate*).ti,ab,hw.

Embase search terms 1. (carb or carbs or carbohydrate*).ti,ab,hw.

Cochrane search terms #1. (carb or carbs or carbohydrate*):ti,ab,kw

F.3.4 Diet 4. In adults with type 1 diabetes, what is the clinical and cost-effectiveness of a diet based on the glycaemic index for optimal diabetic control?

Search constructed by combining the columns in the following table using the AND Boolean operator Population Intervention Study filter used Date parameters Type 1 diabetes High glycaemic index The following filters All available dates [see diet were used in Medline Table 1] and Embase only: RCTs or SRs.

National Clinical Guideline Centre, 2014 123 Type 1 diabetes in adults

Literature search strategies

Medline search terms 1. glycemic index/ 2. exp dietary carbohydrates/ 3. (glyc?emic index or gi).ti,ab. 4. ((carbohydrate? or cho) adj3 (count* or quant* or exchang* or diet* or intake)).ti,ab. 5. chox.ti,ab. 6. or/1-6

Embase search terms 1. glycemic index/ 2. exp dietary carbohydrates/ 3. (glyc?emic index or gi).ti,ab. 4. ((carbohydrate? or cho) adj3 (count* or quant* or exchang* or diet* or intake)).ti,ab. 5. chox.ti,ab. 6. or/1-6

Cochrane search terms #1. MeSH descriptor: [Glycemic Index] this term only #2. MeSH descriptor: [Dietary Carbohydrates] explode all trees #3. (glyc?emic index or GI):ti,ab,kw #4. ((carbohydrate? or CHO) near/3 (count* or quant* or exchang* or diet* or intake)):ti,ab,kw #5. chox:ti,ab #6. {or #1-#5}

F.3.5 HbA1c The same search strategy was used for the following two questions. 5. In adults with type 1 diabetes, what is the optimum target HbA1c level that should be achieved to reduce the risk of complications? 6. In adults with type 1 diabetes, what is optimum frequency of HbA1c monitoring for effective diabetic control?

Search constructed by combining the columns in the following table using the AND Boolean operator Population Intervention Study filter used Date parameters Type 1 diabetes HbA1c The following filters All available dates [see were used in Medline Table 1] and Embase only: RCTs or SRs or observational studies or prognostic studies [Medline and Embase only]

Medline search terms 1. *hemoglobin a, glycosylated/ 2. time factors/ 3. 1 and 2 4. ((HbA1c or hba or ghb or hbaic or a1c or hba1) adj3 (target* or profile* or goal* or timing or time* or frequency or frequent* or regularity or regular* or rate or control* or optimal or optimum or level or levels or concentration* or lower* or rais* or change* or higher or association* or recorded or decreas* or model* or control)).ti,ab.

National Clinical Guideline Centre, 2014 124 Type 1 diabetes in adults

Literature search strategies

5. ((glycosylated or glycated or glycoslated) adj3 (haemoglobin or hemoglobin) adj3 (target* or profile* or goal* or timing or time* or frequency or frequent* or regularity or regular* or rate or control* or optimal or optimum or level or levels or concentration* or lower* or rais* or change* or higher or association* or recorded or decreas* or model* or control)).ti,ab. 6. or/4-5 7. (HbA1c or hba or ghb or hbaic or a1c or hba1).ti,ab. 8. ((glycosylated or glycated or glycoslated) adj3 (haemoglobin or hemoglobin)).ti,ab. 9. or/7-8 10. ((timing or time* or frequency or frequent* or regularity or regular* or rate) adj3 (monitor* or test* or check*)).ti,ab. 11. 9 and 10 12. 3 or 6 or 11

Embase search terms 1. *glycosylated hemoglobin/ 2. time/ 3. 1 and 2 4. ((HbA1c or hba or ghb or hbaic or a1c or hba1) adj3 (target* or profile* or goal* or timing or time* or frequency or frequent* or regularity or regular* or rate or control* or optimal or optimum or level or levels or concentration* or lower* or rais* or change* or higher or association* or recorded or decreas* or model* or control)).ti,ab. 5. ((glycosylated or glycated or glycoslated) adj3 (haemoglobin or hemoglobin) adj3 (target* or profile* or goal* or timing or time* or frequency or frequent* or regularity or regular* or rate or control* or optimal or optimum or level or levels or concentration* or lower* or rais* or change* or higher or association* or recorded or decreas* or model* or control)).ti,ab. 6. or/4-5 7. (HbA1c or hba or ghb or hbaic or a1c or hba1).ti,ab. 8. ((glycosylated or glycated or glycoslated) adj3 (haemoglobin or hemoglobin)).ti,ab. 9. or/7-8 10. ((timing or time* or frequency or frequent* or regularity or regular* or rate) adj3 (monitor* or test* or check*)).ti,ab. 11. 9 and 10 12. 3 or 6 or 11

Cochrane search terms #1. MeSH descriptor: [hemoglobin a, glycosylated] this term only #2. MeSH descriptor: [time factors] this term only #3. #1 and #2 #4. ((HbA1c or hba or ghb or hbaic or a1c or hba1) near/3 (target* or profile* or goal* or timing or time* or frequency or frequent* or regularity or regular* or rate or control* or optimal or optimum or level or levels or concentration* or lower* or rais* or change* or higher or association* or recorded or decreas* or model* or control)):ti,ab,kw #5. ((glycosylated or glycated or glycoslated) near/3 (haemoglobin or hemoglobin) near/3 (target* or profile* or goal* or timing or time* or frequency or frequent* or regularity or regular* or rate or control* or optimal or optimum or level or levels or concentration* or lower* or rais* or change* or higher or association* or recorded or decreas* or model* or control)):ti,ab,kw #6. {or #4-#5} #7. (HbA1c or hba or ghb or hbaic or a1c or hba1):ti,ab,kw #8. ((glycosylated or glycated or glycoslated) near/3 (haemoglobin or hemoglobin)):ti,ab,kw #9. {or #7-#8}

National Clinical Guideline Centre, 2014 125 Type 1 diabetes in adults

Literature search strategies

#10. ((timing or time* or frequency or frequent* or regularity or regular* or rate) near/3 (monitor* or test* or check*)):ti,ab,kw #11. #9 and #10 #12. #3 or #6 or #11

F.3.6 Self-monitoring of blood glucose (SMBG)

The same search strategy was used for the following two questions. 7. In adults with type 1 diabetes, what is optimum timing and frequency to self-monitor blood glucose for effective diabetic control? 8. In adults with type 1 diabetes, what is the optimum glucose target/profile for self-monitoring of blood glucose for effective diabetic control?

Search constructed by combining the columns in the following table using the AND Boolean operator Population Intervention Study filter used Date parameters Type 1 diabetes Self-monitoring of blood None All available dates [see glucose Table 1]

Medline search terms 1. blood glucose/ or (blood adj glucose*).ti,ab. or (blood adj sugar*).ti,ab. 2. self monitor*.ti,ab. 3. 1 and 2 4. blood glucose self-monitoring/ 5. ((home or self) adj6 (blood or glucose) adj6 (monitor* or test*)).ti,ab. 6. (hmbg* or smbg*).ti,ab. 7. ((glyc?emic or glucose) adj2 (target* or profile*)).ti,ab. 8. or/4-7 9. 3 or 8

Embase search terms 1. *glucose blood level/ or (blood adj glucose*).ti,ab. or (blood adj sugar*).ti,ab. 2. self monitor*.ti,ab. 3. 1 or 2 4. *blood glucose monitoring/ 5. ((home or self) adj6 (blood or glucose) adj6 (monitor* or test*)).ti,ab. 6. (smbg* or hmbg*).ti,ab. 7. ((glyc?emic or glucose) adj4 (target* or profile*)).ti,ab. 8. or/4-7 9. 3 or 8

Cochrane search terms #1. MeSH descriptor: [blood glucose] this term only #2. (blood near/1 glucose*):ti,ab,kw #3. (blood near/1 sugar*):ti,ab,kw #4. {or #1-#3} #5. self monitor*:ti,ab,kw #6. #4 and #5 #7. MeSH descriptor: [blood glucose self-monitoring] this term only

National Clinical Guideline Centre, 2014 126 Type 1 diabetes in adults

Literature search strategies

#8. ((home or self) near/6 (blood or glucose) near/6 (monitor* or test*)):ti,ab,kw #9. (smbg* or hmbg*):ti,ab,kw #10. ((glycaemic or glycemic or glucose) near/4 (target* or profile*)):ti,ab,kw #11. {or #7-#10} #12. #6 or #11

F.3.7 Glucose monitoring

The same search strategy was used for the following three questions. 9. In adults with type 1 diabetes, is retrospective continuous glucose monitoring more effective than care without continuous glucose monitoring (with SMBG) for improving diabetic control? 10. In adults with type 1 diabetes, is real-time continuous glucose monitoring more effective than SMBG continuous glucose monitoring for optimum diabetic control? 11. In adults with type 1 diabetes, is continuous real-time monitoring more effective than intermittent real-time monitoring for optimum diabetic control?

Search constructed by combining the columns in the following table using the AND Boolean operator Population Intervention Study filter used Date parameters Type 1 diabetes Glucose monitoring The following filters 2003 - 28 August 2014 were used in Medline [see Table 1] and Embase only: RCTs or SRs or cohort studies [Medline and Embase only]

Medline search terms 1. blood glucose self-monitoring/ 2. ((glucose or continuous or real time or intermittent or retrospective) and monitor*).ti. 3. ((glucose or continuous or real time or intermittent or retrospective) adj5 (monitor* or measure*)).ab. 4. (cgm* or bgm* or smbg*).ti,ab. 5. *hemoglobin a, glycosylated/ and monitor*.ti,ab. 6. or/1-5

Embase search terms 1. blood glucose monitoring/ 2. blood glucose meter/ 3. ((glucose or continuous or real time or intermittent or retrospective) and monitor*).ti. 4. ((glucose or continuous or real time or intermittent or retrospective) adj5 (monitor* or measure*)).ab. 5. (cgm* or bgm* or smbg*).ti,ab. 6. *glycosylated hemoglobin/ and monitor*.ti,ab. 7. or/1-6

Cochrane search terms #1. MeSH descriptor: [blood glucose self-monitoring] explode all trees #2. ((glucose or continuous or "real time" or intermittent or retrospective) and monitor*):ti #3. ((glucose or continuous or “real time” or intermittent or retrospective) near/5 (monitor* or measure*)):ab #4. (cgm* or bgm* or smbg*):ti,ab

National Clinical Guideline Centre, 2014 127 Type 1 diabetes in adults

Literature search strategies

#5. #1 or #2 or #3 or #4

F.3.8 Technologies 12. In adults with type 1 diabetes, what are the benefits of technologies (bolus calculators and downloads) for self-monitoring of blood glucose?

Search constructed by combining the columns in the following table using the AND Boolean operator Population Intervention Study filter used Date parameters Type 1 diabetes Technologies None All available dates [see Table 1]

Medline search terms 1. blood glucose self-monitoring/ 2. ((blood or glucose or glycemic or sugar) adj6 (manage* or monitor* or test* or target* or control*)).ti,ab. 3. (home or self).ti,ab. 4. 2 and 3 5. (hmbg* or smbg* or finger prick*).ti,ab. 6. or/1,4-5 7. ((new* or recent* or edge or latest or state or mobile or emerg* or innovat*) adj3 technolog*).ti,ab. 8. application*.ti,ab. 9. (in check or dafne online or on track or tracker or easy diabetes or ration wizard or calorie counter or diabetes self-management or diabetes diary or diabetes manager or diabetes planner or diabetes log* or glucose companion or diabetes companion or diabetes pilot or diabetes management system or diabetes recorder or bg monitor or my diabetes).ti,ab. 10. or/7-9 11. 6 and 10 12. patient-centered care/mt 13. (smart phone* or smartphone* or iphone* or android or app or apps or mobile phone* or cell phone*).ti,ab. 14. download*.ti,ab. 15. cellular phone/ 16. ((bolus or insulin dose) adj3 (calculator* or wizard*)).ti,ab. 17. (carbs adj2 cals).ti,ab. 18. (glucose buddy or insulin dose calculator pro or ibgstar or mysugr or tactiohealth or diamedic or rapidcalc or diappbetes or diabetes gps or healthsome g for glucose or bant or betabetes or ontrack or glucool diabetes or sidiary or dialog or ifora or glucatrends or d sharp or diabetical or glucowave or track3).ti,ab. 19. or/12-18 20. 11 or 19

Embase search terms 1. *blood glucose monitoring/ 2. ((blood or glucose or glycemic or sugar) adj6 (manage* or monitor* or test* or target* or control*)).ti,ab. 3. (home or self).ti,ab. 4. 2 and 3 5. (hmbg* or smbg* or finger prick*).ti,ab.

National Clinical Guideline Centre, 2014 128 Type 1 diabetes in adults

Literature search strategies

6. or/1,4-5 7. ((new* or recent* or edge or latest or state or mobile or emerg* or innovat*) adj3 technolog*).ti,ab. 8. application*.ti,ab. 9. (in check or dafne online or on track or tracker or easy diabetes or ration wizard or calorie counter or diabetes self-management or diabetes diary or diabetes manager or diabetes planner or diabetes log* or glucose companion or diabetes companion or diabetes pilot or diabetes management system or diabetes recorder or bg monitor or my diabetes).ti,ab. 10. or/7-9 11. 6 and 10 12. (smart phone* or smartphone* or iphone* or android or app or apps or mobile phone* or cell phone*).ti,ab. 13. download*.ti,ab. 14. mobile phone/ 15. ((bolus or insulin dose) adj3 (calculator* or wizard*)).ti,ab. 16. (glucose buddy or insulin dose calculator pro or ibgstar or mysugr or tactiohealth or diamedic or rapidcalc or diappbetes or diabetes gps or healthsome g for glucose or bant or betabetes or ontrack or glucool diabetes or sidiary or dialog or ifora or glucatrends or d sharp or diabetical or glucowave or track3).ti,ab. 17. (carbs adj2 cals).ti,ab. 18. or/12-17 19. 11 or 18

Cochrane search terms #1. MeSH descriptor: [blood glucose self-monitoring] this term only #2. ((blood or glucose or glycemic or sugar) near/6 (manage* or monitor* or test* or target* or control*)):ti,ab,kw #3. (home or self):ti,ab,kw #4. #2 and #3 #5. (hmbg* or smbg* or finger prick*):ti,ab,kw #6. #1 or #4 or #5 #7. ((new* or recent* or edge or latest or state or mobile or emerg* or innovat*) near/3 technolog*):ti,ab,kw #8. application*:ti,ab,kw #9. (in check or dafne online or on track or tracker or easy diabetes or ration wizard or calorie counter or diabetes self-management or diabetes diary or diabetes manager or diabetes planner or diabetes log* or glucose companion or diabetes companion or diabetes pilot or diabetes management system or diabetes recorder or bg monitor or my diabetes):ti,ab,kw #10. {or #7-#9} #11. #6 and #10 #12. (smart phone* or smartphone* or iphone* or android or app or apps or mobile phone* or cell phone*):ti,ab,kw #13. download*:ti,ab,kw #14. MeSH descriptor: [cellular phone] this term only #15. ((bolus or insulin dose) near/3 (calculator* or wizard*)):ti,ab,kw #16. (carbs near/2 cals):ti,ab,kw #17. (glucose buddy or insulin dose calculator pro or ibgstar or mysugr or tactiohealth or diamedic or rapidcalc or diappbetes or diabetes gps or healthsome g for glucose or bant or betabetes or ontrack or glucool diabetes or sidiary or dialog or ifora or glucatrends or d sharp or diabetical

National Clinical Guideline Centre, 2014 129 Type 1 diabetes in adults

Literature search strategies

or glucowave or track3):ti,ab,kw #18. {or #12-#17} #19. #11 or #18

F.3.9 Long-acting/basal insulin

The same search strategy was used for the following two questions. 13. In adults with type 1 diabetes, what are the most effective long-acting insulins (detemir versus degludec versus glargine versus NPH) for optimal diabetic control? 14. In adults with type 1 diabetes, is once daily basal insulin more effective than twice daily basal insulin for optimal diabetic control?

Search constructed by combining the columns in the following table using the AND Boolean operator Population Intervention Study filter used Date parameters Type 1 diabetes Long-acting insulins The following filters 2003 - 28 August 2014 were used in Medline [see Table 1] and Embase only: RCTs and SRs [Medline and Embase only]

Medline search terms 1. insulin, isophane/ 2. insulin, long-acting/ 3. (detemir* or degludec* or glargine*).ti,ab. 4. (tresiba* or lantus* or optisulin* or levemir*).ti,ab. 5. ((isofan* or isophan* or nph or protaphan* or lente) and insulin*).ti,ab. 6. (lente adj2 (hypurin* or insulin*)).ti,ab. 7. (protamin* and (zinc or neutral or hagedorn)).ti,ab. 8. (insuman adj5 basal).ti,ab. 9. (humulin* adj (i or n or r)).ti,ab. 10. (actraphan* or berlinsulin* or insulatard* or monotard* or mixtard * or novolin* or iletin* or umuline* or orgasuline*).ti,ab. 11. or/1-10

Embase search terms 1. long acting insulin/ 2. insulin degludec/ 3. insulin glargine/ 4. insulin detemir/ 5. isophane insulin/ 6. (detemir* or degludec* or glargine*).ti,ab. 7. (tresiba* or lantus* or optisulin* or levemir*).ti,ab. 8. ((isofan* or isophan* or nph or protaphan* or lente) and insulin*).ti,ab. 9. (lente adj2 (hypurin* or insulin*)).ti,ab. 10. (protamin* and (zinc or neutral or hagedorn)).ti,ab. 11. (insuman* adj5 basal).ti,ab. 12. (humulin* adj (i or n or r)).ti,ab. 13. (actraphan* or berlinsulin* or insulatard* or monotard* or mixtard * or novolin* or iletin* or umuline* or orgasuline*).ti,ab.

National Clinical Guideline Centre, 2014 130 Type 1 diabetes in adults

Literature search strategies

14. or/1-13

Cochrane search terms #1. MeSH descriptor: [insulin, long-acting] explode all trees #2. (detemir* or degludec* or glargine*):ti,ab #3. (tresiba* or lantus* or detemir* or optisulin* or levemir*):ti,ab #4. ((isofan* or isophan* or nph or protaphan* or lente) and insulin*):ti,ab #5. (lente near/2 (hypurin or insulin or insulins)):ti,ab #6. (protamin* and (zinc or neutral or hagedorn)):ti,ab #7. (insuman near/5 basal):ti,ab #8. (“humulin i” or “humulin n” or “humulin r”):ti,ab #9. (actraphan* or berlinsulin* or insulatard* or monotard* or mixtard * or novolin* or iletin* or umuline* or orgasuline*):ti,ab #10. #1 or #2 or #3 or #4 or #5 or #6 or #7 or #8 or #9

F.3.10 Rapid-acting insulins 15. In adults with type 1 diabetes, which are the most effective rapid-acting insulins for meal times: analogues versus human (intermediate NPH), for optimal diabetic control?

Search constructed by combining the columns in the following table using the AND Boolean operator Population Intervention Study filter used Date parameters Type 1 diabetes Rapid-acting insulins The following filters 2003 - 28 August 2014 were used in Medline [see Table 1] and Embase only: RCTs or SRs or cohort studies [Medline and Embase only]

Medline search terms 1. postprandial period/ 2. (prandial* or postprandial* or preprandial*).ti,ab. 3. (meal* or premeal or postmeal).ti,ab. 4. (lunch* or dinner* or breakfast*).ti,ab. 5. or/1-4 6. exp insulin,short-acting/ 7. (insulin* and (short or rapid)).ti,ab. 8. (insulin* and (fast adj3 act*)).ti,ab. 9. (aspart or glulisine or lispro).ti,ab. 10. ((porcine or bovine or animal or human) adj3 insulin*).ti,ab. 11. (nph or neutral protamine hagedorn).ti,ab. 12. or/6-11 13. 5 and 12

Embase search terms 1. postprandial state/ 2. (prandial* or postprandial* or preprandial*).ti,ab. 3. (meal* or premeal or postmeal).ti,ab. 4. (lunch* or dinner* or breakfast*).ti,ab. 5. or/1-4

National Clinical Guideline Centre, 2014 131 Type 1 diabetes in adults

Literature search strategies

6. short acting insulin/ 7. insulin aspart/ 8. insulin glulisine/ 9. insulin lispro/ 10. (insulin* and (short or rapid)).ti,ab. 11. (insulin* and (fast adj3 act*)).ti,ab. 12. (aspart or glulisine or lispro).ti,ab. 13. ((porcine or bovine or animal or human) adj3 insulin*).ti,ab. 14. (nph or neutral protamine hagedorn).ti,ab. 15. or/6-14 16. 5 and 15

Cochrane search terms #1. MeSH descriptor: [postprandial period] this term only #2. (prandial* or postprandial* or preprandial*):ti,ab #3. (meal* or premeal or postmeal):ti,ab #4. (lunch* or dinner* or breakfast*):ti,ab #5. #1 or #2 or #3 or #4 #6. MeSH descriptor: [insulin, short-acting] explode all trees #7. (insulin* and (short or rapid)):ti,ab #8. (insulin* and (fast near/3 act*)):ti,ab #9. (aspart or glulisine or lispro):ti,ab #10. ((porcine or bovine or animal or human) near/3 insulin*):ti,ab #11. (nph or “neutral protamine hagedorn”):ti,ab #12. #6 or #7 or #8 or #9 or #10 or #11 #13. #5 and #12

F.3.11 Mixed insulins 16. In adults with type 1 diabetes, what are the most effective mixed insulins (degludec-aspart versus glargine versus NPH) for optimal diabetic control?

Search constructed by combining the columns in the following table using the AND Boolean operator Population Intervention Study filter used Date parameters Type 1 diabetes Mixed insulins The following filters All available dates [see were used in Medline Table 1] and Embase only: RCTs or SRs [Medline and Embase only]

Medline search terms 1. biphasic insulins/ 2. ((biphasic or mix* or premix*) adj4 insulin*).ti,ab. 3. (novomix* or biasp* or iasp* or mixtard* or idegasp*).ti,ab. 4. (degludec* adj3 aspart).ti,ab. 5. (insuman* adj3 comb*).ti,ab. 6. ((humalog* or bd or lispro* or novolog* or aspart* or novolin* or humulin* or isophan* or nph* or neutral protamine hagedorn or glulisine or insuman*) adj4 (mix* or premix* or biphasic or "25" or "30" or "50" or "70" or "75")).ti,ab.

National Clinical Guideline Centre, 2014 132 Type 1 diabetes in adults

Literature search strategies

7. (degludec* adj4 (mix* or premix* or biphasic or "25" or "30" or "50" or "70" or "75")).ti,ab. 8. ((tresiba* or optisulin*) adj4 (mix* or premix* or biphasic or "25" or "30" or "50" or "70" or "75")).ti,ab. 9. ((isofan* or protaphan* or lente) adj4 (mix* or premix* or biphasic or "25" or "30" or "50" or "70" or "75")).ti,ab. 10. ((isofan* or protaphan* or lente) adj4 (mix* or premix* or biphasic or "25" or "30" or "50" or "70" or "75")).ti,ab. 11. ((actraphan* or berlinsulin* or insulatard* or monotard* or iletin* or umuline* or orgasuline*) adj4 (mix* or premix* or biphasic or "25" or "30" or "50" or "70" or "75")).ti,ab. 12. ((hypurin or porcine) adj4 (mix* or premix* or biphasic or "25" or "30" or "50" or "70" or "75")).ti,ab. 13. or/1-12

Embase search terms 1. biphasic insulin/ 2. ((biphasic or mix* or premix*) adj4 insulin*).ti,ab. 3. (novomix* or biasp* or iasp* or mixtard* or idegasp*).ti,ab. 4. (degludec* adj3 aspart).ti,ab. 5. (insuman* adj3 comb*).ti,ab. 6. ((humalog* or bd or lispro* or novolog* or aspart* or novolin* or humulin* or isophan* or nph* or neutral protamine hagedorn or glulisine or insuman*) adj4 (mix* or premix* or biphasic or "25" or "30" or "50" or "70" or "75")).ti,ab. 7. (degludec* adj4 (mix* or premix* or biphasic or "25" or "30" or "50" or "70" or "75")).ti,ab. 8. ((tresiba* or optisulin*) adj4 (mix* or premix* or biphasic or "25" or "30" or "50" or "70" or "75")).ti,ab. 9. ((isofan* or protaphan* or lente) adj4 (mix* or premix* or biphasic or "25" or "30" or "50" or "70" or "75")).ti,ab. 10. (protamin* adj4 (mix* or premix* or biphasic or "25" or "30" or "50" or "70" or "75")).ti,ab. 11. ((actraphan* or berlinsulin* or insulatard* or monotard* or iletin* or umuline* or orgasuline*) adj4 (mix* or premix* or biphasic or "25" or "30" or "50" or "70" or "75")).ti,ab. 12. ((hypurin or porcine) adj4 (mix* or premix* or biphasic or "25" or "30" or "50" or "70" or "75")).ti,ab. 13. or/1-12

Cochrane search terms #1. MeSH descriptor: [biphasic insulins] this term only #2. ((biphasic or mix* or premix*) near/4 insulin*):ti,ab,kw #3. (novomix* or biasp* or iasp* or mixtard* or idegasp*):ti,ab,kw #4. ((humalog* or lispro* or novolog* or aspart* or novolin* or humulin* or isophan* or nph* or neutral protamine hagedorn or glulisine or insuman*) near/4 (mix* or premix* or biphasic or "25" or "30" or "50" or "70" or "75")):ti,ab,kw #5. (degludec* near/3 aspart):ti,ab,kw #6. (insuman* near/3 comb*):ti,ab,kw #7. (degludec* near/4 (mix* or premix* or biphasic or "25" or "30" or "50" or "70" or "75")):ti,ab,kw #8. ((tresiba* or optisulin*) near/4 (mix* or premix* or biphasic or "25" or "30" or "50" or "70" or "75")):ti,ab,kw #9. ((isofan* or protaphan* or lente) near/4 (mix* or premix* or biphasic or "25" or "30" or "50" or "70" or "75")):ti,ab,kw #10. ((isofan* or protaphan* or lente) near/4 (mix* or premix* or biphasic or "25" or "30" or "50"

National Clinical Guideline Centre, 2014 133 Type 1 diabetes in adults

Literature search strategies

or "70" or "75")):ti,ab,kw #11. ((actraphan* or berlinsulin* or insulatard* or monotard* or iletin* or umuline* or orgasuline*) near/4 (mix* or premix* or biphasic or "25" or "30" or "50" or "70" or "75")):ti,ab,kw #12. ((hypurin or porcine) near/4 (mix* or premix* or biphasic or "25" or "30" or "50" or "70" or "75")):ti,ab,kw #13. #1 or #2 or #3 or #4 or #5 or #6 or #7 or #8 or #9 or #10 or #11 or #12

F.3.12 Metformin and GLP-1 agonists 17. In adults with type 1 diabetes, are metformin (with or without insulin), or GLP1-agonists (with or without insulin) as effective as insulin alone for optimal diabetic control?

Search constructed by combining the columns in the following table using the AND Boolean operator Population Intervention Study filter used Date parameters Type 1 diabetes Metformin or exenatide The following filters 2003 - 28 August 2014 or liraglutide or were used in Medline [see Table 1] pramlintide and Embase only: RCTs or SRs [Medline and Embase only]

Medline search terms 1. metformin/ 2. *glucagon-like peptide 1/ 3. (metformin* or exenatide* or exendin 4 or pramlintide* or liraglutide*).ti,ab. 4. (byetta* or bydureon* or symlin* or victoza* or glucophage* or riomet* or fortamet* or glumetza* or bolamyn* or glucient* or metabet*).ti,ab. 5. or/1-4

Embase search terms 1. metformin/ 2. *glucagon like peptide 1/ 3. exendin 4/ 4. pramlintide/ 5. liraglutide/ 6. (metformin* or exenatide* or exendin 4 or pramlintide* or liraglutide*).ti,ab. 7. (byetta* or bydureon* or symlin* or victoza* or glucophage* or riomet* or fortamet* or glumetza* or bolamyn* or glucient* or metabet*).ti,ab. 8. or/1-7

Cochrane search terms #1. MeSH descriptor: [metformin] this term only #2. (metformin* or exenatide* or "exendin 4" or pramlintide* or liraglutide*):ti,ab #3. (byetta* or bydureon* or symlin* or victoza* or glucophage* or riomet* or fortamet* or glumetza* or bolamyn* or glucient* or metabet*):ti,ab #4. #1 or #2 or #3

F.3.13 Glucose injections

The same search strategy was used for the following two questions. 18. In adults with type 1 diabetes, what is the optimum needle length for insulin delivery? 19. In adults with type 1 diabetes, what is the optimum injection site and rotation for insulin delivery?

National Clinical Guideline Centre, 2014 134 Type 1 diabetes in adults

Literature search strategies

Search constructed by combining the columns in the following table using the AND Boolean operator Population Intervention Study filter used Date parameters Type 1 diabetes Needles (length, site, The following filters 2003 - 28 August 2014 rotation) were used in Medline [see Table 1] and Embase only: RCTs or SRs or cohort studies [Medline and Embase only]

Medline search terms 1. needles/ 2. syringes/ 3. (needle* or syringe* or pen or pens or microneedle* or vial*).ti,ab. 4. (inject* and (length or gauge* or width or technique* or site* or rotat*)).ti,ab. 5. (autopen or clikstar or humapen or novopen or opticlik or optipen or flexpen or solostar or kwikpen or flextouch or miriopen).ti,ab. 6. or/1-5

Embase search terms 1. insulin pen/ 2. needle/ 3. syringe/ 4. (needle* or syringe* or pen or pens or microneedle* or vial*).ti,ab. 5. (inject* and (length or gauge* or width or technique* or site* or rotat*)).ti,ab. 6. (autopen or clikstar or humapen or novopen or opticlik or optipen or flexpen or solostar or kwikpen or flextouch or miriopen).ti,ab. 7. or/1-6

Cochrane search terms #1. MeSH descriptor: [needles] this term only #2. MeSH descriptor: [syringes] this term only #3. (pen or pens or microneedle* or vial* or syringe* or needle*):ti,ab #4. (inject* and (length or gauge* or width or technique* or rotat* or site*)):ti,ab #5. (autopen or clikstar or humapen or novopen or opticlik or optipen or flexpen or solostar or kwikpen or flextouch or miriopen):ti,ab #6. #1 or #2 or #3 or #4 or #5

F.3.14 Pancreas transplantation

For the following question specific sources were used to obtain the data on referral criteria. These data sources are detailed in the review protocol in Appendix C. 20. Which adults with type 1 diabetes are most suitable to be considered for a pancreas transplant, or pancreatic islet cell transplantation?

F.3.15 Hypoglycaemic awareness

The same search strategy was used for the following two questions. 21. In adults with type 1 diabetes, how is impaired awareness of hypoglycaemia best identified and quantified?

National Clinical Guideline Centre, 2014 135 Type 1 diabetes in adults

Literature search strategies

22. In adults with type 1 diabetes and impaired awareness of hypoglycaemia, what is the most effective strategy for recovering hypoglycaemia awareness?

Search constructed by combining the columns in the following table using the AND Boolean operator Population Intervention Study filter used Date parameters Type 1 diabetes Hypoglycaemic None All available dates [see awareness Table 1]

Medline search terms 1. exp hypoglycemia/ 2. (hypoglycaem* or hypoglycem* or hypo glycem* or hypo glycaem*).ti,ab. 3. (low adj3 blood adj3 (sugar or glucose)).ti,ab. 4. ((glycemic or glycaemic) adj control).ti,ab. 5. or/1-4 6. awareness/ 7. (aware* or unaware*).ti,ab. 8. (symptom* adj3 (detect* or recogni*)).ti,ab. 9. (gold or clarke or ryan or danish or pedersen or bjergaard).ti,ab. 10. or/6-9 11. 5 and 10 12. ((glycaem* or glycem* or hypoglycem* or hypoglycaem*) adj2 (impair* or recogni* or aware* or unaware* or identif* or quantif* or problem* or autonomic or iatrogenic)).ti,ab. 13. (iah or haaf).ti,ab. 14. ((hypo or beta or hypoglycaem* or hypoglycem*) adj3 (burden or compass or score* or scale* or method* or algorithm* or recurrent)).ti,ab. 15. or/12-14 16. 11 or 15

Embase search terms 1. exp hypoglycemia/ 2. (hypoglycaem* or hypoglycem* or hypo glycem* or hypo glycaem*).ti,ab. 3. (low adj3 blood adj3 (sugar or glucose)).ti,ab. 4. ((glycemic or glycaemic) adj control).ti,ab. 5. or/1-4 6. awareness/ 7. (aware* or unaware*).ti,ab. 8. (symptom* adj3 (detect* or recogni*)).ti,ab. 9. (gold or clarke or ryan or danish or pedersen or bjergaard).ti,ab. 10. or/6-9 11. 5 and 10 12. ((glycaem* or glycem* or hypoglycem* or hypoglycaem*) adj2 (impair* or recogni* or aware* or unaware* or identif* or quantif* or problem* or autonomic or iatrogenic)).ti,ab. 13. (iah or haaf).ti,ab. 14. ((hypo or beta or hypoglycaem* or hypoglycem*) adj3 (burden or compass or score* or scale* or method* or algorithm* or recurrent)).ti,ab. 15. or/12-14 16. 11 or 15

Cochrane search terms

National Clinical Guideline Centre, 2014 136 Type 1 diabetes in adults

Literature search strategies

#1. MeSH descriptor: [hypoglycemia] explode all trees #2. (hypoglycaem* or hypoglycem* or hypo glycem* or hypo glycaem*):ti,ab #3. (low near/3 blood near/3 (sugar or glucose)):ti,ab #4. ((glycemic or glycaemic) near/1 control):ti,ab #5. {or #1-#4} #6. MeSH descriptor: [awareness] this term only #7. (aware* or unaware*):ti,ab #8. (symptom* near/3 (detect* or recogni*)):ti,ab #9. (gold or clarke or ryan or danish or pedersen or bjergaard):ti,ab #10. {or #6-#9} #11. #5 and #10 #12. ((glycaem* or glycem* or hypoglycem* or hypoglycaem*) near/2 (impair* or recogni* or aware* or unaware* or identif* or quantif* or problem* or autonomic or iatrogenic)):ti,ab #13. (haaf or iah):ti,ab #14. ((hypo or beta or hypoglycaem* or hypoglycem*) near/3 (burden or compass or score* or scale* or method* or algorithm* or recurrent)):ti,ab #15. {or #12-#14} #16. #11 or #15

F.3.16 Ketone monitoring

The same search strategy was used for the following two questions. 23. In adults with type 1 diabetes (including atypical ketosis-prone diabetes), does patient self- monitoring of blood (and urine) ketones reduce the incidence of diabetic ketoacidosis and hospital admissions? 24. In adults with type 1 diabetes does in-patient monitoring of blood ketones by the healthcare professional reduce the length of hospital stay, exposure to IV insulin and the development of in- hospital complications: a. in patients with suspected diabetic ketoacidosis? b. in patients admitted with diabetic ketoacidosis and/or those that get it in hospital.

Search constructed by combining the columns in the following table using the AND Boolean operator Population Intervention Study filter used Date parameters Type 1 diabetes Ketone monitoring None All available dates [see Table 1]

Medline search terms 1. ketone bodies/ 2. 3-hydroxybutyric acid/ 3. (hydroxybutyrate or hydroxybutyric).ti,ab. 4. (ketone* or ketosis or ketogenesis).ti,ab. 5. or/1-4 6. monitoring, physiologic/ 7. blood glucose self-monitoring/ 8. point-of-care systems/ 9. predictive value of tests/ 10. "sensitivity and specificity"/ 11. exp reagent kits, diagnostic/

National Clinical Guideline Centre, 2014 137 Type 1 diabetes in adults

Literature search strategies

12. (test* or monitor* or screen* or measure* or detect* or diagnos* or predict* or sensitivity or specificity).ti,ab. 13. or/6-12 14. 5 and 13

Embase search terms 1. 3 hydroxybutyric acid/ 2. ketone/ 3. ketone body/ 4. ketogenesis/ 5. ketonuria/ 6. (hydroxybutyrate or hydroxybutyric).ti,ab. 7. (ketone* or ketosis or ketogenesis).ti,ab. 8. or/1-7 9. patient monitoring/ 10. monitoring/ 11. blood glucose monitoring/ 12. diagnostic test accuracy study/ 13. diagnostic value/ 14. glucose blood level/ 15. predictive value/ 16. "sensitivity and specificity"/ 17. point of care testing/ 18. test strip/ 19. (test* or monitor* or screen* or measure* or detect* or diagnos* or predict* or sensitivity or specificity).ti,ab. 20. or/9-19 21. 8 and 20

Cochrane search terms #1. MeSH descriptor ketone bodies, this term only #2. MeSH descriptor ketosis, this term only #3. MeSH descriptor 3-hydroxybutyric acid, this term only #4. (hydroxybutyrate or hydroxybutyric):ti,ab #5. (ketone* or ketosis or ketogenesis):ti,ab #6. (#1 or #2 or #3 or #4 or #5) #7. MeSH descriptor monitoring, physiologic, this term only #8. MeSH descriptor blood glucose self-monitoring, this term only #9. MeSH descriptor point-of-care systems, this term only #10. MeSH descriptor predictive value of tests, this term only #11. MeSH descriptor sensitivity and specificity, this term only #12. MeSH descriptor reagent kits, diagnostic explode all trees #13. (test* or monitor* or screen* or measure* or detect* or diagnos* or predict* or sensitivity or specificity):ti,ab #14. (#7 or #8 or #9 or #10 or #11 or #12 or #13) #15. (#6 and #14)

National Clinical Guideline Centre, 2014 138 Type 1 diabetes in adults

Literature search strategies

F.3.17 Aspirin for prevention of cardiovascular events

For the following question a broad diabetes population was used, hence the searches for this question are reproduced in full below. 25. In adults with type 1 diabetes, is aspirin an effective anti-platelet agent for the primary prevention of cardiovascular events?

Search constructed by combining the columns in the following table using the AND Boolean operator Population Intervention Study filter used Date parameters Diabetes Aspirin The following filters 2003 - 28 August 2014 were used in Medline [see Table 1] and Embase only: RCTs or SRs [Medline and Embase only]

Medline search terms 1. diabet*.ti,ab,hw. 2. ((children or adolescen* or school* or infant* or teenage* or paediatric* or pediatric*) not (adult* or onset)).ti. 3. (pregnan* or gestation* or foot or feet or optic or eye* or retin* or ocular).ti. 4. 1 not (2 or 3) 5. (acetylsalicylic acid or aspirin).ti,ab,hw. 6. 4 and 5 7. letter/ 8. editorial/ 9. news/ 10. exp historical article/ 11. anecdotes as topic/ 12. comment/ 13. case report/ 14. (letter or comment*).ti. 15. or/7-14 16. 15 not (randomized controlled trial/ or random*.ti,ab.) 17. animals/ not humans/ 18. exp animals, laboratory/ 19. exp animal experimentation/ 20. exp models, animal/ 21. exp rodentia/ 22. (rat or rats or mouse or mice).ti. 23. or/16-22 24. 6 not 23

Embase search terms 1. diabet*.ti,ab,hw. 2. ((children or adolescen* or school* or infant* or teenage* or paediatric* or pediatric*) not (adult* or onset)).ti. 3. (pregnan* or gestation* or foot or feet or optic or eye* or retin* or ocular).ti. 4. 1 not (2 or 3) 5. letter.pt. or letter/

National Clinical Guideline Centre, 2014 139 Type 1 diabetes in adults

Literature search strategies

6. note.pt. 7. editorial.pt. 8. case report/ or case study/ 9. (letter or comment*).ti. 10. or/5-9 11. 10 not (randomized controlled trial/ or random*.ti,ab.) 12. animal/ not human/ 13. nonhuman/ 14. exp animal experiment/ 15. exp experimental animal/ 16. animal model/ 17. exp rodent/ 18. (rat or rats or mouse or mice).ti. 19. or/11-18 20. 4 not 19 21. aspirin.ti,ab,hw. 22. *acetylsalicylic acid/ 23. 20 and (21 or 22)

Cochrane search terms #1. MeSH descriptor diabetes mellitus explode all trees #2. diabet*:ti,ab #3. (#1 or #2) #4. ((children or adolescen* or school* or infant* or teenage* or paediatric* or pediatric*) not (adult* or onset)):ti #5. (pregnan* or gestation* or foot or feet or optic or eye* or retin* or ocular):ti #6. (#3 and not ( #4 or #5 )) #7. MeSH descriptor aspirin, this term only #8. ("acetylsalicylic acid" or aspirin):ti,ab #9. (#6 and ( #7 or #8 ))

F.3.18 IV insulin 26. In adults with type 1 diabetes who have been admitted to hospital (elective and emergency), what is the most effective intravenous insulin dose-adjustment devices and regimens for optimal diabetic control?

Search constructed by combining the columns in the following table using the AND Boolean operator Population Intervention Study filter used Date parameters Type 1 diabetes IV insulin The following filters All available dates [see were used in Medline Table 1] and Embase only: RCTs or SRs or Observational studies [Medline and Embase only]

Medline search terms 1. exp insulins/ 2. exp administration, intravenous/

National Clinical Guideline Centre, 2014 140 Type 1 diabetes in adults

Literature search strategies

3. 1 and 2 4. (ivit or vriii).ti,ab. 5. ((intravenous* or iv) adj6 insulin*).ti,ab. 6. ((infusion* adj6 insulin*) not subcutaneous).ti,ab. 7. ((actraphan* or aspart or berlinsulin* or biasp* or degludec* or detemir* or glargine* or glulisine or humalog* or humulin* or hypurin or iasp* or idegasp* or iletin* or insulatard* or insuman or isofan* or isophan* or lantus* or lente or levemir* or lispro or mixtard* or monotard* or novolin* or novolog* or novomix* or optisulin* or orgasuline* or protaphan* or protamin* or tresiba* or umuline*) adj6 (intravenous* or iv or infusion*)).ti,ab. 8. or/4-7 9. 3 and 8

Embase search terms 1. exp insulin derivative/ 2. intravenous drug administration/ 3. 1 and 2 4. exp insulin derivative/iv [intravenous drug administration] 5. (ivit or vriii).ti,ab. 6. ((intravenous* or iv) adj6 insulin*).ti,ab. 7. ((infusion* adj6 insulin*) not subcutaneous).ti,ab. 8. ((actraphan* or aspart or berlinsulin* or biasp* or degludec* or detemir* or glargine* or glulisine or humalog* or humulin* or hypurin or iasp* or idegasp* or iletin* or insulatard* or insuman or isofan* or isophan* or lantus* or lente or levemir* or lispro or mixtard* or monotard* or novolin* or novolog* or novomix* or optisulin* or orgasuline* or protaphan* or protamin* or tresiba* or umuline*) adj6 (intravenous* or iv or infusion*)).ti,ab. 9. or/4-8 10. 3 or 9

Cochrane search terms #1. MeSH descriptor: [insulins] explode all trees #2. MeSH descriptor: [administration, intravenous] explode all trees #3. #1 and #2 #4. (ivit or vriii):ti,ab #5. ((intravenous* or iv) near/6 insulin*):ti,ab #6. ((infusion* near/6 insulin*) not subcutaneous):ti,ab #7. ((actraphan* or aspart or berlinsulin* or biasp* or degludec* or detemir* or glargine* or glulisine or humalog* or humulin* or hypurin or iasp* or idegasp* or iletin* or insulatard* or insuman or isofan* or isophan* or lantus* or lente or levemir* or lispro or mixtard* or monotard* or novolin* or novolog* or novomix* or optisulin* or orgasuline* or protaphan* or protamin* or tresiba* or umuline*) near/6 (intravenous* or iv or infusion*)):ti,ab #8. {or #4-#7} #9. #3 or #8

F.3.19 Gastroparesis

For the following question a broad diabetes population was used, hence the searches for this question are reproduced in full below. 27. In adults with type 1 diabetes, what is the most effective treatment for gastroparesis?

Search constructed by combining the columns in the following table using the AND Boolean operator Population Intervention Study filter used Date parameters

National Clinical Guideline Centre, 2014 141 Type 1 diabetes in adults

Literature search strategies

Population Intervention Study filter used Date parameters Diabetes Gastroparesis The following filters 2003 - 28 August 2014 were used in Medline [see Table 1] and Embase only: RCTs, SRs and cohort studies [Medline and Embase only]

Medline search terms 1. diabet*.ti,ab,hw. 2. ((children or adolescen* or school* or infant* or teenage* or paediatric* or pediatric*) not (adult* or onset)).ti. 3. (pregnan* or gestation*).ti. 4. 1 not (2 or 3) 5. letter/ 6. editorial/ 7. news/ 8. exp historical article/ 9. anecdotes as topic/ 10. comment/ 11. case report/ 12. (letter or comment*).ti. 13. or/5-12 14. 13 not (randomized controlled trial/ or random*.ti,ab.) 15. animals/ not humans/ 16. exp animals, laboratory/ 17. exp animal experimentation/ 18. exp models, animal/ 19. exp rodentia/ 20. (rat or rats or mouse or mice).ti. 21. or/14-20 22. 4 not 21 23. (gastropare* or gastropleg*).ti,ab,hw. 24. exp gastrointestinal motility/ 25. ((gastr* or stomach) adj3 (empty* or motility or dysmotility or transit or passage or aton* or pares* or paralys*)).ti,ab. 26. or/23-25 27. 22 and 26

Embase search terms 1. diabet*.ti,ab,hw. 2. ((children or adolescen* or school* or infant* or teenage* or paediatric* or pediatric*) not (adult* or onset)).ti. 3. (pregnan* or gestation*).ti. 4. 1 not (2 or 3) 5. letter.pt. or letter/ 6. note.pt. 7. editorial.pt.

National Clinical Guideline Centre, 2014 142 Type 1 diabetes in adults

Literature search strategies

8. case report/ or case study/ 9. (letter or comment*).ti. 10. or/5-9 11. 10 not (randomized controlled trial/ or random*.ti,ab.) 12. animal/ not human/ 13. nonhuman/ 14. exp animal experiment/ 15. exp experimental animal/ 16. animal model/ 17. exp rodent/ 18. (rat or rats or mouse or mice).ti. 19. or/11-18 20. 4 not 19 21. stomach emptying/ 22. stomach paresis/ 23. gastrointestinal motility/ 24. gastrointestinal transit/ 25. migrating myoelectric complex/ 26. peristalsis/ 27. (gastropare* or gastropleg*).ti,ab. 28. ((gastr* or stomach) adj3 (empty* or motility or dysmotility or transit or passage or aton* or pares* or paralys*)).ti,ab. 29. or/21-28 30. 20 and 29

Cochrane search terms #1. MeSH descriptor diabetes mellitus explode all trees #2. diabet*:ti,ab #3. (#1 or #2) #4. ((children or adolescen* or school* or infant* or teenage* or paediatric* or pediatric*) not (adult* or onset)):ti #5. (pregnan* or gestation*):ti #6. (#3 and not ( #4 or #5 )) #7. MeSH descriptor gastrointestinal motility explode all trees #8. MeSH descriptor gastroparesis, this term only #9. (gastropare* or gastropleg*):ti,ab #10. ((gastr* or stomach) near/3 (empty* or motility or dysmotility or transit or passage or aton* or pares* or paralys*)):ti,ab #11. (#6 and (#7 or #8 or #9 or #10 ))

F.3.20 Thyroid disease 28. How should adults with type 1 diabetes be monitored for thyroid disease, and how frequently?

Search constructed by combining the columns in the following table using the AND Boolean operator Population Exposure Study filter used Date parameters Type 1 diabetes Thyroid disease The following filters All available dates [see were used in Medline Table 1]

National Clinical Guideline Centre, 2014 143 Type 1 diabetes in adults

Literature search strategies

Population Exposure Study filter used Date parameters and Embase only: RCTs or SRs or cohort studies or prognostic [Medline and Embase only]

Medline search terms 1. exp thyroid diseases/ not exp thyroid neoplasms/ 2. exp thyroid hormones/ or thyroid function tests/ or thyrotropin-releasing hormone/ 3. (thyroid* or hypothyroid* or hyperthyroid* or hyperthyrox* or myxedem* or myxoedem* or thyrotoxic*).ti,ab. 4. (hashimoto* or ords or euthyroid* or grave* or nodular struma or plummer* or hashitoxicosis or goitre or goiter or thyroglossal duct*).ti,ab. 5. (peroxidase or antithyroglobulin*).ti,ab. 6. (protein-bound iodine or pbi).ti,ab. 7. (thyroxine or thyrotropin or triiodothyronine or diiodothyronine or iodothyronine or dextrothyroxine or diiodotyrosine or monoiodotyrosine or iodotyrosine).ti,ab. 8. or/1-7

Embase search terms 1. exp thyroid disease/ not exp thyroid tumor/ 2. thyroid hormone/ or thyroid function test/ or protirelin/ 3. (thyroid* or hypothyroid* or hyperthyroid* or hyperthyrox* or myxedem* or myxoedem* or thyrotoxic*).ti,ab. 4. (hashimoto* or ords or euthyroid* or grave* or nodular struma or plummer* or hashitoxicosis or goitre or goiter or thyroglossal duct*).ti,ab. 5. (peroxidase or antithyroglobulin*).ti,ab. 6. (protein-bound iodine or pbi).ti,ab. 7. (thyroxine or thyrotropin or triiodothyronine or diiodothyronine or iodothyronine or dextrothyroxine or diiodotyrosine or monoiodotyrosine or iodotyrosine).ti,ab. 8. or/1-7

Cochrane search terms #1. MeSH descriptor: [thyroid diseases] explode all trees #2. MeSH descriptor: [thyroid neoplasms] explode all trees #3. #1 not #2 #4. MeSH descriptor: [thyroid hormones] explode all trees #5. MeSH descriptor: [thyroid function tests] this term only #6. MeSH descriptor: [thyrotropin-releasing hormone] this term only #7. (thyroid* or hypothyroid* or hyperthyroid* or hyperthyrox* or myxedem* or myxoedem* or thyrotoxic*):ti,ab #8. (hashimoto* or ords or euthyroid* or grave* or nodular struma or plummer* or hashitoxicosis or goitre or goiter or thyroglossal duct*):ti,ab #9. (peroxidase or antithyroglobulin*):ti,ab #10. (protein-bound iodine or pbi):ti,ab #11. (thyroxine or thyrotropin or triiodothyronine or diiodothyronine or iodothyronine or dextrothyroxine or diiodotyrosine or monoiodotyrosine or iodotyrosine):ti,ab #12. {or #4-#11} #13. #3 or #12

National Clinical Guideline Centre, 2014 144 Type 1 diabetes in adults

Literature search strategies

F.3.21 Neuropathy 29. In adults with type 1 diabetes, what is the most effective treatment for insulin-induced neuropathy?

Search constructed by combining the columns in the following table using the AND Boolean operator Population Intervention Study filter used Date parameters Type 1 diabetes Neuropathy The following filters All available dates [see were used in Medline Table 1] and Embase only: RCTs or SRs or observational studies [Medline and Embase only]

Medline search terms 1. diabetic neuropathies/ 2. polyneuropathies/ 3. mononeuropathies/ 4. (neuropath* or polyneuropath* or mononeuropath* or neuriti* or polyneuriti* or mononeuriti*).ti,ab. 5. third nerv*.ti,ab. 6. ((diabet* or insulin*) adj3 (amyotroph* or neuralgi*)).ti,ab. 7. or/1-6

Embase search terms 1. diabetic neuropathy/ 2. polyneuropathy/ 3. mononeuropathy/ 4. autonomic neuropathy/ 5. peripheral neuropathy/ 6. (neuropath* or polyneuropath* or mononeuropath* or neuriti* or polyneuriti* or mononeuriti*).ti,ab. 7. third nerv*.ti,ab. 8. ((diabet* or insulin*) adj3 (amyotroph* or neuralgi*)).ti,ab. 9. or/1-8

Cochrane search terms #1. MeSH descriptor: [diabetic neuropathies] this term only #2. MeSH descriptor: [polyneuropathies] this term only #3. MeSH descriptor: [mononeuropathies] this term only #4. (neuropath* or polyneuropath* or mononeuropath* or neuriti* or polyneuriti* or mononeuriti*):ti,ab #5. ("third nerve"):ti,ab #6. ((diabet* or insulin*) near/3 (amyotroph* or neuralgi*)):ti,ab #7. {or #1-#6}

F.4 Economic searches

F.4.1 Economic reviews

Economic searches were run in Medline and Embase by combining the standard population with the economic filter (F.1.6) and limiting by a date range of 2009 onwards. Economic searches were

National Clinical Guideline Centre, 2014 145 Type 1 diabetes in adults

Literature search strategies

executed in the HEED and CRD (NHS EED and HTA) databases by running a population with a date limit of 2003 onwards. Search terms are given below.

Search constructed by combining the columns in the following table using the AND Boolean operator Population Study filter used Date parameters Type 1 diabetes The following filters were used in 2009 – 28 August 2014 (Medline Medline and Embase only: and Embase) Economic [only Embase and 2003 - 28 August 2014 (NHS EED, Medline] HTA and HEED)

CRD (NHS EED, HTA) search terms #1. MeSH descriptor diabetes mellitus, type 1 explode all trees #2. MeSH descriptor diabetic ketoacidosis #3. (diabet*) #4. MeSH descriptor blood glucose #5. ((blood near1 (glucose* or sugar))) #6. (((glucose or sugar or glycemic) near2 (control* or monitor* or level* or measur* or test*))) #7. #1 or #2 or #3 or #4 or #5 or #6

HEED search terms 1. ax=diabet* 2. ax=type 1 3. ax=type i 4. ax=insulin dependent 5. ax=autoimmune 6. ax=sudden onset 7. ax=juvenile 8. ax=childhood 9. cs=1 and (2 or 3 or 4 or 5 or 6 or 7 or 8)

F.4.2 Quality of life reviews

Quality of life (QOL) searches were run in Medline and Embase by combining the standard population with the QOL filter (F.1.7) and limiting to a date range of 2003 onwards.

Search constructed by combining the columns in the following table using the AND Boolean operator Population Study filter used Date parameters Type 1 diabetes The following filters were used in 2003 - 28 August 2014 Medline and Embase only: QOL

National Clinical Guideline Centre, 2014 146 Type 1 diabetes in adults

References

References

1 National Institute for Health and Clinical Excellence. The guidelines manual. London: National Institute for Health and Clinical Excellence; 2012. Available from: http://publications.nice.org.uk/the-guidelines-manual-pmg6/

National Clinical Guideline Centre, 2014 147